This page documents production updates to Vertex AI.
You can periodically check this page for announcements about new or updated
features, bug fixes, known issues, and deprecated functionality.

See also:

- [Colab Enterprise release notes](https://docs.cloud.google.com/colab/docs/release-notes.md.txt)
- [Generative AI on Vertex AI release notes](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/release-notes.md.txt)
- [Vertex AI Workbench release notes](https://docs.cloud.google.com/vertex-ai/docs/workbench/release-notes)

You can see the latest product updates for all of Google Cloud on the
[Google Cloud](https://docs.cloud.google.com/release-notes.md.txt) page, browse and filter all release notes in the
[Google Cloud console](https://console.cloud.google.com/release-notes),
or programmatically access release notes in
[BigQuery](https://console.cloud.google.com/bigquery?p=bigquery-public-data&d=google_cloud_release_notes&t=release_notes&page=table).

To get the latest product updates delivered to you, add the URL of this page to your
[feed
reader](https://wikipedia.org/wiki/Comparison_of_feed_aggregators), or add the
[feed URL](https://docs.cloud.google.com/feeds/vertex-ai-product-group-release-notes.xml) directly.

## November 10, 2026

**Agent Platform Workbench** v2 Feature **Antigravity CLI in Agent Platform Workbench instances**

[Preview](https://cloud.google.com/products#product-launch-stages): The
Antigravity CLI is available in Agent Platform Workbench instances that use a
Debian 12 (`workbench-instances-2603`) or later image. The Antigravity CLI is
the terminal surface of Google Antigravity, an AI agent that works in a
terminal. New instances have it turned on by default. You can use the
Antigravity CLI to do tasks like the following:

- Create, explain, edit, check, and run notebooks.
- Create a Jupyter kernel with the packages that you need.
- Run shell commands and Google Cloud commands, such as `gcloud` and `bq`.
- Write the code for a notebook cell from a description, with the `%%agy` cell magic.

The Antigravity CLI asks for your approval before it edits a file or runs a
command. To turn it off for an instance, set the `enable-antigravity` metadata
key to `false`. To get started, see [Use the Antigravity
CLI](https://docs.cloud.google.com/gemini-enterprise-agent-platform/notebooks/workbench/instances/antigravity-cli).

## November 09, 2026

**Colab Enterprise** Feature Python 3.13 is now available in Colab Enterprise and is the latest
Python version. Runtime templates that are configured to use the latest Python
version, including default runtime templates, now create runtimes that use
Python 3.13. Existing runtimes keep their current Python version. For more
information, see [Python versions](https://docs.cloud.google.com/colab/docs/runtimes.md.txt#python).

## November 02, 2026

**Agent Platform Workbench** v2 Change

### 20261101.00_p0 Release

Change Installed latest packages from upstream dependencies.
Security Upgraded the container host to the latest Container-Optimized OS LTS release, rebasing onto the newest operating system, Go, and containerd security fixes.
Security Updated the CA certificate configuration.
Change

### 20261101.00_p0 Release

Change Installed latest packages from upstream dependencies.
Security Upgraded the container host to the latest Container-Optimized OS LTS release, rebasing onto the newest operating system, Go, and containerd security fixes.
Security Updated the CA certificate configuration.

## November 01, 2026

**Agent Platform Workbench** v2 Change

### 20261101-2230-rc0 Release

Change

### 20261101-2230-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Security Updated the CA certificate configuration.
Security Upgraded the container host to the latest Container-Optimized OS LTS release, rebasing onto the newest operating system, Go, and containerd security fixes.
Security Updated the CA certificate configuration.
Security Upgraded the container host to the latest Container-Optimized OS LTS release, rebasing onto the newest operating system, Go, and containerd security fixes.
Change

### 20261101-2130-rc0 Release

Change Installed latest packages from upstream dependencies.
Security Updated the CA certificate configuration.
Fixed Changes to the idle shutdown timeout now take effect on running instances without a restart. Previously, an instance could still be stopped after the old timeout until it was restarted.
Fixed Fixed an issue where temporary metadata server errors could delay idle shutdown or stop it from working until the instance was restarted.
Change

### M155 Release

Change Installed latest packages from upstream dependencies.
Security Updated the CA certificate configuration.
Fixed Changes to the idle shutdown timeout now take effect on running instances without a restart. Previously, an instance could still be stopped after the old timeout until it was restarted.
Fixed Fixed an issue where temporary metadata server errors could delay idle shutdown or stop it from working until the instance was restarted.

## October 26, 2026

**Agent Platform Workbench** v2 Change

### 20261025.00_p0 Release

Change Installed latest packages from upstream dependencies.
Security Updated certifi, pip, and conda ca-certificates to the latest versions.
Change

### 20261025.00_p0 Release

Change Installed latest packages from upstream dependencies.
Security Addressed multiple reported security vulnerabilities in bundled packages.
Security Updated certifi, pip, and conda ca-certificates to the latest versions.
Feature **Custom organization policies for Agent Platform Workbench instances**

You can now use custom organization policies to allow or deny the creation and
update of Agent Platform Workbench instances based on their configuration. Custom
constraints for the `notebooks.googleapis.com/Instance` resource are enforced
when you create or update an instance, and support dry-run mode. This feature is
now generally available
([GA](https://cloud.google.com/products#product-launch-stages)). For more
information, see [Create custom organization policy
constraints](https://docs.cloud.google.com/gemini-enterprise-agent-platform/notebooks/workbench/instances/custom-constraints).

## October 25, 2026

**Agent Platform Workbench** v2 Change

### 20261025-2230-rc0 Release

Change

### 20261025-2230-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Security Updated certifi, pip, and conda ca-certificates to the latest versions.
Fixed Fixed dataset and table search in the BigQuery panel, which sometimes failed to return results. Root cause is due to Datacatalog API being sunsetted in favor of Dataplex API.
Fixed Fixed an issue where Jupyter Server terminal configuration could overwrite the Dataproc plugin server extension setting or fail to apply the terminal disable setting on container startup.
Security Addressed multiple reported security vulnerabilities in bundled packages.
Security Updated certifi, pip, and conda ca-certificates to the latest versions.
Fixed Fixed an issue where Jupyter Server terminal configuration could overwrite the Dataproc plugin server extension setting or fail to apply the terminal disable setting on container startup.
Security Addressed multiple reported security vulnerabilities in bundled packages.
Change

### 20261025-2130-rc0 Release

Change Installed latest packages from upstream dependencies.

## October 20, 2026

**Agent Platform Workbench** v2 Change

### M154 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed dataset and table search in the BigQuery panel, which sometimes failed to return results. Root cause is due to Datacatalog API being sunsetted in favor of Dataplex API.

## October 19, 2026

**Agent Platform Workbench** v2 Change

### 20261018.00_p0 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed an issue where Jupyter Server terminal configuration could overwrite the Dataproc plugin server extension setting or fail to apply the terminal disable setting on container startup.
Change

### 20261018.00_p0 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed an issue where Jupyter Server terminal configuration could overwrite the Dataproc plugin server extension setting or fail to apply the terminal disable setting on container startup.

## October 13, 2026

**Agent Platform Workbench** v2 Change

### 20261013-2230-rc0 Release

Change

### 20261013-2230-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.

## October 12, 2026

**Agent Platform Workbench** v2 Change

### 20261011.00_p0 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed an issue where gcloud and gsutil failed to load with an AttributeError in OpenSSL.crypto due to a version mismatch between pyOpenSSL and cryptography.
Change

### 20261011.00_p0 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed an issue where gcloud and gsutil failed to load with an AttributeError in OpenSSL.crypto due to a version mismatch between pyOpenSSL and cryptography.

## October 11, 2026

**Agent Platform Workbench** v2 Change

### 20261011-2230-rc0 Release

Change

### 20261011-2230-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Fixed Fixed an issue where gcloud and gsutil failed to load with an AttributeError in OpenSSL.crypto due to a version mismatch between pyOpenSSL and cryptography.
Fixed Fixed an issue where gcloud and gsutil failed to load with an AttributeError in OpenSSL.crypto due to a version mismatch between pyOpenSSL and cryptography.
Change

### 20261011-2130-rc0 Release

Change Installed latest packages from upstream dependencies.

## October 06, 2026

**Agent Platform Workbench** v2 Change

### M153 Release

Change Installed latest packages from upstream dependencies.

## October 05, 2026

**Agent Platform Workbench** v2 Change

### 20261004.00_p0 Release

Change Installed latest packages from upstream dependencies.
Change The BigQuery JupyterLab plugin is now enabled independently of the Dataproc plugin: it is enabled by default and can be turned off with the new disable-bigquery instance metadata key, while disable-mixer now controls only the Dataproc plugin.
Change

### 20261004.00_p0 Release

Change Installed latest packages from upstream dependencies.

## October 04, 2026

**Agent Platform Workbench** v2 Change

### 20261004-2230-rc0 Release

Change

### 20261004-2230-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Change The BigQuery JupyterLab plugin is now enabled independently of the Dataproc plugin: it is enabled by default and can be turned off with the new disable-bigquery instance metadata key, while disable-mixer now controls only the Dataproc plugin.
Security Bulk security patch remediating Critical and High-severity CVEs in the Agent Platform Workbench custom container images.
Security Bulk security patch remediating Critical and High-severity CVEs in the Agent Platform Workbench custom container images.
Change

### 20261004-2157-rc0 Release

Change Installed latest packages from upstream dependencies.
Fixed Preinstall CUDA drivers for Pre-Turing GPUs such as V100s.
Change The BigQuery JupyterLab plugin is now enabled independently of the Dataproc plugin: it is enabled by default and can be turned off with the new disable-bigquery instance metadata key, while disable-mixer now controls only the Dataproc plugin.
Change

### M152 Release

Change Installed latest packages from upstream dependencies.
Change The BigQuery JupyterLab plugin is now enabled independently of the Dataproc plugin: it is enabled by default and can be turned off with the new disable-bigquery instance metadata key, while disable-mixer now controls only the Dataproc plugin.

## September 28, 2026

**Agent Platform Workbench** v2 Change

### 20260927.00_p0 Release

Change Installed latest packages from upstream dependencies.
Security Bulk security patch remediating Critical and High-severity CVEs in the Workbench custom container images.
Change

### 20260927.00_p0 Release

Change Installed latest packages from upstream dependencies.
Security Bulk security patch remediating Critical and High-severity CVEs in the Workbench custom container images.
**Colab Enterprise** Feature You can now use the Google Cloud console to set an existing runtime template
as the default runtime template for a project and region, with or without
accelerators. New default runtimes in that project and region use the
template's configuration. For more information, see
[Set a runtime template as the default runtime template](https://docs.cloud.google.com/colab/docs/runtimes.md.txt#set-default-runtime-template).

## September 27, 2026

**Agent Platform Workbench** v2 Change

### 20260927-2130-rc0 Release

Change Installed latest packages from upstream dependencies.
Change

### M151 Release

Change Installed latest packages from upstream dependencies.
Feature JupyterLab now forwards client-side logs (console errors, uncaught exceptions, unhandled promise rejections, and failed network requests) to the instance backend, where they surface in Cloud Logging for easier debugging.
Change

### 20260927-1408-rc0 Release

Change

### 20260927-1408-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.

## September 21, 2026

**Agent Platform Workbench** v2 Change

### 20260920.00_p0 Release

Change Installed latest packages from upstream dependencies.
Feature JupyterLab now forwards client-side logs (console errors, uncaught exceptions, unhandled promise rejections, and failed network requests) to the instance backend, where they surface in Cloud Logging for easier debugging.
Change

### 20260920.00_p0 Release

Change Installed latest packages from upstream dependencies.
Feature JupyterLab now forwards client-side logs (console errors, uncaught exceptions, unhandled promise rejections, and failed network requests) to the instance backend, where they surface in Cloud Logging for easier debugging.

## September 18, 2026

**Agent Platform Workbench** v2 Change

### 20260918-2230-rc0 Release

Change

### 20260918-2230-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Feature JupyterLab now forwards client-side logs (console errors, uncaught exceptions, unhandled promise rejections, and failed network requests) to the instance backend, where they surface in Cloud Logging for easier debugging.
Feature JupyterLab now forwards client-side logs (console errors, uncaught exceptions, unhandled promise rejections, and failed network requests) to the instance backend, where they surface in Cloud Logging for easier debugging.
Change

### 20260918-2130-rc0 Release

Change Installed latest packages from upstream dependencies.
Feature JupyterLab now forwards client-side logs (console errors, uncaught exceptions, unhandled promise rejections, and failed network requests) to the instance backend, where they surface in Cloud Logging for easier debugging.

## September 14, 2026

**Agent Platform Workbench** v2 Change

### M150 Release

Change Installed latest packages from upstream dependencies.

## September 13, 2026

**Agent Platform Workbench** v2 Change

### 20260911.00_p0 Release

Change

### 20260913.00_p0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Fixed Fixed the %%bigquery notebook cell magic, which returned an error instead of query results in JupyterLab 4.
Fixed Fixed the %%bigquery notebook cell magic, which returned an error instead of query results in JupyterLab 4.
Change

### 20260913-2230-rc0 Release

Change

### 20260913-2230-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Fixed Fixed an issue where the `notebook-disable-nbconvert` metadata flag was ignored in custom containers.
Change The obsolete
`google-cloud-sdk` transitional package is no longer installed. The Google Cloud
CLI itself is unchanged; it was already provided by the `google-cloud-cli`
package.
Fixed Fixed the %%bigquery notebook cell magic, which returned an error instead of query results in JupyterLab 4.
Fixed Fixed an issue where the `notebook-disable-nbconvert` metadata flag was ignored in custom containers.
Fixed Fixed the %%bigquery notebook cell magic, which returned an error instead of query results in JupyterLab 4.
Change

### 20260913-2130-rc0 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed the %%bigquery notebook cell magic, which returned an error instead of query results in JupyterLab 4.
Change

### M149 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed the %%bigquery notebook cell magic, which returned an error instead of query results in JupyterLab 4.

## September 09, 2026

**Agent Platform Workbench** v2 Feature **New configuration options for scheduled notebook runs**

When you schedule a notebook run from an Agent Platform Workbench instance, you
can now configure the following:

- **Execution identity.** You can choose your own user account instead of a service account. The notebook runs as you, so Application Default Credentials inside the notebook resolve to your identity and the run accesses only the resources you can access. If you have not yet granted consent, the scheduler provides a link for authorizing access.
- **Shielded VM options.** To meet workload requirements that mandate verified boot integrity, you can enable Shielded VM options, including Secure Boot, on the compute that runs the notebook. For more information, see [Modifying Shielded VM
  options](https://cloud.google.com/compute/docs/instances/modifying-shielded-vm).
- **Runtime environment.** In addition to the default environment, you can specify a VM image or a custom container, so that scheduled runs use the same dependencies as your interactive environment.

Ensure that `scheduler-jupyter-plugin` is at version 0.1.8 or later for these
features to be accessible. For more information, see [Schedule a notebook
run](https://docs.cloud.google.com/gemini-enterprise-agent-platform/notebooks/workbench/instances/schedule-notebook-run-quickstart).

## September 08, 2026

**Agent Platform Workbench** v2 Fixed **Scheduled upgrade metadata is validated**

The value of the `notebook-upgrade-schedule` metadata key is now validated when
you create or update an Agent Platform Workbench instance. The value must be a
single-line [unix-cron
format](https://man7.org/linux/man-pages/man5/crontab.5.html) schedule. For more
information, see [Manage features through
metadata](https://docs.cloud.google.com/gemini-enterprise-agent-platform/notebooks/workbench/instances/manage-metadata#keys).

## August 30, 2026

**Agent Platform Workbench** v2 Change

### 20260830-2330-rc0 Release

Change

### 20260831.01_p0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Fixed The terminal and file browser are now more responsive when working in a Cloud Storage bucket that is mounted through the file browser. Frequent background checks for non-existent files are now cached instead of repeatedly querying Cloud Storage, which previously could make actions such as listing files or running terminal commands take several seconds.
Fixed Cloud Storage buckets that you mount through the file browser are now automatically re-mounted after the instance is restarted or is stopped and started. Previously the mounted folder could be left behind as an empty, unusable directory that had to be manually removed and re-mounted.
Change The obsolete
google-cloud-sdk transitional package is no longer installed. The Google Cloud
CLI itself is unchanged; it was already provided by the google-cloud-cli
package.
Change

### 20260830-2230-rc0 Release

Change

### 20260830-2230-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Fixed Cloud Storage buckets that you mount through the file browser are now automatically re-mounted after the instance is restarted or is stopped and started. Previously the mounted folder could be left behind as an empty, unusable directory that had to be manually removed and re-mounted.
Fixed The terminal and file browser are now more responsive when working in a Cloud Storage bucket that is mounted through the file browser. Frequent background checks for non-existent files are now cached instead of repeatedly querying Cloud Storage, which previously could make actions such as listing files or running terminal commands take several seconds.
Change

### 20260830-2154-rc1 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed an issue where a notebook's cells and their outputs could be unexpectedly erased. The automatic reload of an open notebook (which keeps it in sync with changes made to its file on disk) is now restricted to run only while the Gemini CLI is in use, so notebooks are no longer overwritten at other times.
Fixed Cloud Storage buckets that you mount through the file browser are now automatically re-mounted after the instance is restarted or is stopped and started. Previously the mounted folder could be left behind as an empty, unusable directory that had to be manually removed and re-mounted.
Fixed The terminal and file browser are now more responsive when working in a Cloud Storage bucket that is mounted through the file browser. Frequent background checks for non-existent files are now cached instead of repeatedly querying Cloud Storage, which previously could make actions such as listing files or running terminal commands take several seconds.
Change Agent Platform Workbench instances internal agents now honor custom CA certificates installed on the host OS (e.g. via a custom VM image), fixing TLS certificate verification failures when Google API traffic is routed through a customer-managed proxy.
Change

### M148 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed an issue where a notebook's cells and their outputs could be unexpectedly erased. The automatic reload of an open notebook (which keeps it in sync with changes made to its file on disk) is now restricted to run only while the Gemini CLI is in use, so notebooks are no longer overwritten at other times.
Fixed The terminal and file browser are now more responsive when working in a Cloud Storage bucket that is mounted through the file browser. Frequent background checks for non-existent files are now cached instead of repeatedly querying Cloud Storage, which previously could make actions such as listing files or running terminal commands take several seconds.
Fixed Cloud Storage buckets that you mount through the file browser are now automatically re-mounted after the instance is restarted or is stopped and started. Previously the mounted folder could be left behind as an empty, unusable directory that had to be manually removed and re-mounted.
Change Agent Platform Workbench instances internal agents now honor custom CA certificates installed on the host OS (e.g. via a custom VM image), fixing TLS certificate verification failures when Google API traffic is routed through a customer-managed proxy.

## August 23, 2026

**Agent Platform Workbench** v2 Change

### 20260823-2330-rc0 Release

Change

### 20260823-2330-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Change

### 20260823-2230-rc0 Release

Change

### 20260823-2230-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Change

### 20260823-2130-rc0 Release

Fixed Scheduled notebook executions now report their final status when the execution user's credentials stop working part way through a run, instead of continuing until the execution timeout.
Change Installed latest packages from upstream dependencies.
Change

### M147 Release

Fixed Scheduled notebook executions now report their final status when the execution user's credentials stop working part way through a run, instead of continuing until the execution timeout.
Change Installed latest packages from upstream dependencies.

## August 16, 2026

**Agent Platform Workbench** v2 Change

### 20260816-2330-rc0 Release

Change

### 20260816-2330-rc0 Release

Security Updated the bundled Ruby gems rexml and net-imap to patched versions, addressing known vulnerabilities.
Security Updated aiohttp, joblib and cryptography to patched versions, addressing known vulnerabilities including CVE-2022-21797 and CVE-2025-69223.
Change Installed latest packages from upstream dependencies.
Change Updated the CUDA base image from 12.8.1 to 12.9.2 (CUDA 12.9, cuDNN 9.10). This is a minor CUDA 12 update, binary compatible with the previous image, and also addresses known vulnerabilities in a bundled NVIDIA profiler component.
Security Updated the bundled Ruby gems rexml and net-imap to patched versions, addressing known vulnerabilities.
Deprecated Removed the JupyterLab 3 environment from the Python 3.12 custom container; JupyterLab 4 is now the only JupyterLab environment and is always used. The Python 3.10 images are unaffected.
Security Updated aiohttp, joblib and cryptography to patched versions, addressing known vulnerabilities including CVE-2022-21797 and CVE-2025-69223.
Change Installed latest packages from upstream dependencies.
Change

### 20260816-2230-rc0 Release

Change

### 20260816-2230-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Change

### 20260816-2130-rc0 Release

Change Installed latest packages from upstream dependencies.
Change

### M146 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed the Git panel's grayed out buttons, which were disabled due to an issue with the Jupyter Lab's Git plugin introduced in version 0.54.0.

## August 09, 2026

**Agent Platform Workbench** v2 Change Installed latest packages from upstream dependencies.
Change

### 20260809-2330-rc0 Release

Change Installed latest packages from upstream dependencies.
Change

### 20260809-2330-rc0 Release

Fixed Fixed the Git panel's grayed out buttons which were disabled due to an issue with the Jupyter Lab's Git plugin introduced in version 0.54.0.
Change

### 20260809-2230-rc0 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed the Git panel's grayed out buttons which were disabled due to an issue with the Jupyter Lab's Git plugin introduced in version 0.54.0.
Change

### 20260809-2230-rc0 Release

Change Installed latest packages from upstream dependencies.
Change

### 20260809-2130-rc0 Release

Change Updated the NVIDIA GPU driver on Workbench Debian 12 images from 580.65.06 to 580.126.20 for compatibility with the Debian 12 6.1.0-52 kernel.
Change Installed latest packages from upstream dependencies.
Fixed Fixed the Git panel's grayed out buttons which were disabled due to an issue with the Jupyter Lab's Git plugin introduced in version 0.54.0.

## August 02, 2026

**Agent Platform Workbench** v2 Change

### 20260802-2330-rc0 Release

Change

### 20260802-2330-rc0 Release

Fixed Fixed issue with JupyterLab UI silently reverting file changes and interrupting kernels due to Gemini CLI's auto reload extension.
Change The JupyterLab last-active and auto-reload extensions are now enabled only when Gemini CLI is configured, preventing them from affecting the native JupyterLab UI otherwise.
Change Installed latest packages from upstream dependencies.
Fixed Fixed issue with JupyterLab UI silently reverting file changes and interrupting kernels due to Gemini CLI's auto reload extension.
Change The JupyterLab last-active and auto-reload extensions are now enabled only when Gemini CLI is configured, preventing them from affecting the native JupyterLab UI otherwise.
Change Installed latest packages from upstream dependencies.
Fixed Fixed an error that could prevent creating new `micromamba` environments at runtime due to package cache permissions.
Fixed Fixed an error that could prevent creating new `micromamba` environments at runtime due to package cache permissions.
Change

### 20260802-2230-rc0 Release

Change

### 20260802-2230-rc0 Release

Fixed Fixed issue with JupyterLab UI silently reverting file changes and interrupting kernels due to Gemini CLI's auto reload extension.
Change The JupyterLab last-active and auto-reload extensions are now enabled only when Gemini CLI is configured, preventing them from affecting the native JupyterLab UI otherwise.
Change Installed latest packages from upstream dependencies.
Fixed Fixed issue with JupyterLab UI silently reverting file changes and interrupting kernels due to Gemini CLI's auto reload extension.
Change The JupyterLab last-active and auto-reload extensions are now enabled only when Gemini CLI is configured, preventing them from affecting the native JupyterLab UI otherwise.
Change Installed latest packages from upstream dependencies.

## July 30, 2026

**Agent Platform Workbench** v2 Change

### M145 Release

Change

### 20260730-2130-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Fixed Fixed issue with JupyterLab UI silently reverting file changes and interrupting kernels due to Gemini CLI's auto reload extension.
Change The JupyterLab last-active and auto-reload extensions are now enabled only when Gemini CLI is configured, preventing them from affecting the native JupyterLab UI otherwise.
Fixed Fixed the JupyterLab Git extension's Pull and Push buttons being disabled by pinning `jupyterlab-git` to 0.53.0.
Fixed Fixed the JupyterLab Git extension's Pull and Push buttons being disabled by pinning `jupyterlab-git` to 0.53.0.
Fixed Fixed an error that could prevent creating new `micromamba` environments at runtime due to package cache permissions.
Fixed Fixed an error that could prevent creating new `micromamba` environments at runtime due to package cache permissions.
Fixed Workbench internal agents now trust custom and enterprise CA certificates installed on the host operating system, fixing TLS certificate verification failures for connections routed through the instance proxy.
Fixed Workbench internal agents now trust custom and enterprise CA certificates installed on the host operating system, fixing TLS certificate verification failures for connections routed through the instance proxy.
Fixed Files created by the post-startup script are now owned by the `jupyter` user, and `git safe.directory` is configured so that root-owned repositories continue to work.
Fixed Files created by the post-startup script are now owned by the `jupyter` user, and `git safe.directory` is configured so that root-owned repositories continue to work.
Fixed Fixed issue with JupyterLab UI silently reverting file changes and interrupting kernels due to Gemini CLI's auto reload extension.
Change The JupyterLab last-active and auto-reload extensions are now enabled only when Gemini CLI is configured, preventing them from affecting the native JupyterLab UI otherwise.
Change The JupyterLab last-active and auto-reload extensions are now enabled only when Gemini CLI is configured, preventing them from affecting the native JupyterLab UI otherwise.

## July 13, 2026

**Agent Platform Workbench** v2 Feature **Agent Platform Workbench image release**

The following Agent Platform Workbench instances image releases are available:

- **20260712-2130-rc0 (`workbench-instances-2603` - Debian 12)**
  - Installed latest packages from upstream dependencies.
  - Fixed broken cupy installation.
- **M144 (`workbench-instances` - Debian 11)**
  - Installed latest packages from upstream dependencies.
  - Fixed a race condition that could cause JupyterLab to be unreachable (HTTP 524) on GPU instances.
Change

### M144 Release

Change

### 20260712-2130-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Fixed Fixed a race condition that could cause JupyterLab to be unreachable (HTTP 524) on GPU instances.
Fixed Fixed broken cupy installation.
Feature **Secure Boot is compatible with GPUs**

You can now enable Secure Boot on Agent Platform Workbench instances that have a
GPU attached. Secure Boot with GPUs is supported on the `workbench-instances-2603`
VM image and the `workbench-container-2606` custom container, which include a
Secure Boot-signed NVIDIA GPU driver so the driver loads under Secure Boot. For
more information, see [Create an
instance](https://docs.cloud.google.com/gemini-enterprise-agent-platform/notebooks/workbench/instances/create).
Feature **Secure Boot is compatible with GPUs**

You can now enable Secure Boot on Agent Platform Workbench instances that have a
GPU attached. Secure Boot with GPUs is supported on the `workbench-instances-2603`
VM image and the `workbench-container-2606` custom container, which include a
Secure Boot-signed NVIDIA GPU driver so the driver loads under Secure Boot. For
more information, see [Create an
instance](https://docs.cloud.google.com/gemini-enterprise-agent-platform/notebooks/workbench/instances/create).
Feature **Secure Boot is compatible with GPUs**

You can now enable Secure Boot on Agent Platform Workbench instances that have a
GPU attached. Secure Boot with GPUs is supported on the `workbench-instances-2603`
VM image and the `workbench-container-2606` custom container, which include a
Secure Boot-signed NVIDIA GPU driver so the driver loads under Secure Boot. For
more information, see [Create an
instance](https://docs.cloud.google.com/gemini-enterprise-agent-platform/notebooks/workbench/instances/create).
**Colab Enterprise** Security A Missing Authorization vulnerability was discovered in repositories in
BigQuery, Dataform, and Colab Enterprise. An authenticated attacker could
potentially escalate permissions and perform cross-tenant repository takeover.
For more information, see the
[GCP-2026-047](https://docs.cloud.google.com/support/bulletins.md.txt#gcp-2026-047) security bulletin.

## July 06, 2026

**Agent Platform Workbench** v2 Feature **Agent Platform Workbench image release**

The following Agent Platform Workbench instances image release is available:

- **20260701-2130-rc0 (`workbench-instances-2603` - Debian 12)**
  - Installed latest packages from upstream dependencies.
  - Fixed a race condition that could cause JupyterLab to be unreachable (HTTP 524) on GPU instances.
  - Fixed an issue where long-running requests (for example, streaming or long-poll connections) could be terminated after about 60 seconds.
Change

### 20260701-2130-rc0 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed a race condition that could cause JupyterLab to be unreachable (HTTP 524) on GPU instances.
Fixed Fixed an issue where long-running requests (for example, streaming or long-poll connections) could be terminated after about 60 seconds.

## June 30, 2026

**Agent Platform Workbench** v2 Feature **Python 3.12 base containers for Agent Platform Workbench custom containers**

You can build [custom containers](https://docs.cloud.google.com/gemini-enterprise-agent-platform/notebooks/workbench/instances/create-custom-container.md.txt)
for Agent Platform Workbench instances using Python 3.12 base containers, in addition
to the default Python 3.10 base containers. The Python 3.12 standard and slim
base containers are available at the following URIs:

- `us-docker.pkg.dev/workbench-images/gcr.io/workbench-container-2606:latest`
- `us-docker.pkg.dev/workbench-images/gcr.io/workbench-container-slim-2606:latest`
Feature **Python 3.12 base containers for Agent Platform Workbench custom containers**

You can build [custom containers](https://docs.cloud.google.com/gemini-enterprise-agent-platform/notebooks/workbench/instances/create-custom-container.md.txt)
for Agent Platform Workbench instances using Python 3.12 base containers, in addition
to the default Python 3.10 base containers. The Python 3.12 standard and slim
base containers are available at the following URIs:

- `us-docker.pkg.dev/workbench-images/gcr.io/workbench-container-2606:latest`
- `us-docker.pkg.dev/workbench-images/gcr.io/workbench-container-slim-2606:latest`
Feature **Python 3.12 base containers for Agent Platform Workbench custom containers**

You can build [custom containers](https://docs.cloud.google.com/gemini-enterprise-agent-platform/notebooks/workbench/instances/create-custom-container.md.txt)
for Agent Platform Workbench instances using Python 3.12 base containers, in addition
to the default Python 3.10 base containers. The Python 3.12 standard and slim
base containers are available at the following URIs:

- `us-docker.pkg.dev/workbench-images/gcr.io/workbench-container-2606:latest`
- `us-docker.pkg.dev/workbench-images/gcr.io/workbench-container-slim-2606:latest`

## June 29, 2026

**Agent Platform Workbench** v2 Feature **Agent Platform Workbench image release**

The following Agent Platform Workbench instances image release is available:

- **20260628-2130-rc0 (`workbench-instances-2603` - Debian 12)**
  - Installed latest packages from upstream dependencies.
  - Installed TensorFlow and PyTorch packages to the default kernel.
Change

### 20260628-2130-rc0 Release

Change Installed latest packages from upstream dependencies.
Feature Installed TensorFlow and PyTorch packages to the default kernel.

## June 24, 2026

**Agent Platform Workbench** v2 Feature **Agent Platform Workbench image release**

The following Agent Platform Workbench instances image release is available:

- **20260624-1604-rc0 (`workbench-instances-2603` - Debian 12)**
  - Installed latest packages from upstream dependencies.
  - Fixed an issue where notebook kernels could become unavailable if the Dataproc plugin failed to load.
Change

### 20260624-1604-rc0 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed an issue where notebook kernels could become unavailable if the Dataproc plugin failed to load.

## June 23, 2026

**Agent Platform Workbench** v2 Feature **Agent Platform Workbench image release**

The following Agent Platform Workbench instances image releases are available:

- **20260622-2130-rc0 (`workbench-instances-2603` - Debian 12)**
  - Installed latest packages from upstream dependencies.
  - Fixed a duplicate "Python 3 (ipykernel)" kernel appearing in the launcher.
- **M143 (`workbench-instances` - Debian 11)**
  - Installed latest packages from upstream dependencies.
Change

### M143 Release

Change

### 20260622-2130-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Fixed Fixed a duplicate "Python 3 (ipykernel)" kernel appearing in the launcher.

## June 18, 2026

**Colab Enterprise** Feature Colab Enterprise is now available in the following regions:

- Berlin, Germany (`europe-west10`)
- Delhi, India (`asia-south2`)
- Paris, France (`europe-west9`)

For more information, see
[Colab Enterprise locations](https://docs.cloud.google.com/colab/docs/locations.md.txt).

## June 01, 2026

**Agent Platform Workbench** v2 Feature **Agent Platform Workbench image release**

The following Agent Platform Workbench instances image release is available:

- **20260531-2130-rc0 (`workbench-instances-2603` - Debian 12)**
  - Installed latest packages from upstream dependencies.
Change

### 20260531-2130-rc0 Release

Change Installed latest packages from upstream dependencies.

## May 28, 2026

**Agent Platform Workbench** v2 Feature **Agent Platform Workbench image release**

The following Agent Platform Workbench instances image release is available:

- **M142 (`workbench-instances` - Debian 11)**
  - Installed latest packages from upstream dependencies.
  - Fixed pip availability in the base environment.
  - Hardened JupyterLab authentication configuration.
Change

### M142 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed pip availability in the base environment.
Change Hardened JupyterLab authentication configuration.

## May 26, 2026

**Colab Enterprise** Feature **Data Science Agent**

[Generally available](https://cloud.google.com/products#product-launch-stages):
Use the Data Science Agent to automate exploratory data analysis,
perform machine learning tasks, and deliver insights from within
a Colab Enterprise notebook. To get started, see
[Use the Data Science Agent](https://docs.cloud.google.com/colab/docs/use-data-science-agent.md.txt).
**Generative AI on Vertex AI** v1 Deprecated **Vertex AI Extensions deprecation**

Vertex AI Extensions is deprecated and will be shut down after November 26, 2026.
We recommend
[migrating to Agent Platform](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/extensions/migrate.md.txt)
to avoid service disruption.

## May 25, 2026

**Agent Platform Workbench** v2 Feature **Agent Platform Workbench image release**

The following Agent Platform Workbench instances image release is available:

- **20260524-2130-rc0 (`workbench-instances-2603` - Debian 12)**
  - Installed latest packages from upstream dependencies.
  - Fixed a Python 3.12 compatibility issue in the JupyterLab server configuration.
  - Hardened JupyterLab authentication configuration.
Change

### 20260524-2130-rc0 Release

Change Installed latest packages from upstream dependencies.
Fixed Fixed a Python 3.12 compatibility issue in the JupyterLab server configuration.
Change Hardened JupyterLab authentication configuration.

## May 11, 2026

**Agent Platform Workbench** v2 Feature **Agent Platform Workbench image release**

The following Agent Platform Workbench instances image release is available:

- **20260510-2130-rc0 (`workbench-instances-2603` - Debian 12)**
  - Installed latest packages from upstream dependencies.
Change

### 20260510-2130-rc0 Release

Change Installed latest packages from upstream dependencies.

## May 04, 2026

**Agent Platform Workbench** v2 Feature **Agent Platform Workbench image release**

The following Agent Platform Workbench instances image release is available:

- **20260503-2130-rc0 (`workbench-instances-2603` - Debian 12)**
  - Installed latest packages from upstream dependencies.
  - Improved metadata server resolution reliability on Debian 12 by using the recommended metadata.google.internal name.
Change

### 20260503-2130-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Improved metadata server resolution reliability on Debian 12 by using the recommended metadata.google.internal name.

## April 27, 2026

**Agent Platform Workbench** v2 Feature **Agent Platform Workbench image release**

The following Agent Platform Workbench instances image release is available:

- **20260426-2130-rc0 (`workbench-instances-2603` - Debian 12)**
  - Installed latest packages from upstream dependencies.
Change

### 20260426-2130-rc0 Release

Change Installed latest packages from upstream dependencies.

## April 20, 2026

**Agent Platform Workbench** v2 Feature **Agent Platform Workbench image release**

The following Agent Platform Workbench instances image releases are available:

- **20260419-2130-rc0 (`workbench-instances-2603` - Debian 12)**
  - Installed latest packages from upstream dependencies.
  - Fixed intermittent connectivity failures caused by proxy TLS certificate rotation.
  - The post-startup script now re-runs after an instance auto-upgrade.
- **M141 (`workbench-instances` - Debian 11)**
  - Installed latest packages from upstream dependencies.
  - Fixed intermittent connectivity failures caused by proxy TLS certificate rotation.
  - The post-startup script now re-runs after an instance auto-upgrade.
Change

### M141 Release

Change

### 20260419-2130-rc0 Release

Change Installed latest packages from upstream dependencies.
Change Installed latest packages from upstream dependencies.
Fixed Fixed intermittent connectivity failures caused by proxy TLS certificate rotation.
Fixed Fixed intermittent connectivity failures caused by proxy TLS certificate rotation.
Change The post-startup script now re-runs after an instance auto-upgrade.
Change The post-startup script now re-runs after an instance auto-upgrade.

## April 17, 2026

**Generative AI on Vertex AI** v1 Feature **RAG Cross Corpus Retrieval**

RAG Cross Corpus Retrieval is available in [public preview](https://cloud.google.com/products#product-launch-stages). This feature allows you to retrieve relevant contexts or generate answers from multiple RAG corpora simultaneously using the `AsyncRetrieveContexts` and `AskContexts` APIs.

For more information, see [RAG Cross Corpus Retrieval](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/rag-engine/cross-corpus-retrieval).

## April 14, 2026

**Generative AI on Vertex AI** v1 Feature **Anthropic's Claude Opus 4.7**

[Claude Opus 4.7](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/claude/opus-4-7)
is available in Model Garden.

## April 13, 2026

**Colab Enterprise** Feature **Visualization cells**

[Generally available](https://cloud.google.com/products#product-launch-stages):
You can use visualization cells to generate interactive and editable
visualizations from within a Colab Enterprise notebook. You can
configure the chart type, aggregation, colors, labels, and other aspects of the
visualization to help you explore data and discover insights. For more
information, see [Use visualization cells](https://docs.cloud.google.com/colab/docs/visualization-cells.md.txt).

## April 10, 2026

**Colab Enterprise** Feature **SQL cells**

[Generally available](https://cloud.google.com/products#product-launch-stages):
You can use SQL cells to write, edit, and run SQL queries directly from your
Colab Enterprise notebooks. For more information, see
[Use SQL cells](https://docs.cloud.google.com/colab/docs/sql-cells.md.txt).

## April 06, 2026

**Generative AI on Vertex AI** v1 Feature **Metadata search for RAG Engine**

Use schema-based metadata search in Vertex AI RAG Engine.
You can define a metadata schema for a corpus, attach metadata to files within
that corpus, and use this metadata to filter contexts during retrieval.
For more information, see
[Filter with metadata search](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/rag-engine/use-metadata-search).

## April 03, 2026

**Generative AI on Vertex AI** v1 Feature **Gemma 4 26B A4B IT** is available as an experimental launch in Model Garden. This is an open model built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input and generating text output.
Gemma 4 26B A4B IT is available as a managed API in Model Garden. To learn more, see
[Gemma 4 26B A4B IT](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/google/gemma-4-26b-a4b-it).
Feature **Vertex AI RAG Engine Serverless mode**

Vertex AI RAG Engine Serverless mode is now available in [public
preview](https://cloud.google.com/products#product-launch-stages). Serverless
mode provides a fully managed database for storing RAG resources that abstracts
away database provisioning and scaling. You can seamlessly switch between
Serverless mode and Spanner mode, which provides dedicated, isolated database
instances.

For more information, see the following:

- [Deployment modes in Vertex AI RAG Engine](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/rag-engine/deployment-modes)
- [Serverless mode](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/rag-engine/serverless-mode)
- [Managing Spanner mode](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/rag-engine/spanner-mode)
- [Switching between modes](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/rag-engine/switching-modes)

## April 02, 2026

**Generative AI on Vertex AI** v1 Feature **Veo 3.1 Lite**

Veo 3.1 Lite is available in [public
preview](https://cloud.google.com/products#product-launch-stages). This release
is our most cost-efficient Veo on Vertex AI model.

For more information, see [3.1 Lite
Generate](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/veo/3-1-generate#3.1-lite-generate-001)
Announcement **Gemini 2.5 model retirement dates updated**

The retirement dates for Gemini 2.5 Pro, Gemini 2.5 Flash-Lite,
and Gemini 2.5 Flash have been updated to October 16, 2026. For more information,
see [Model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/model-versions).

## March 30, 2026

**Agent Platform Workbench** v2 Feature **M140 release**

The M140 release of Vertex AI Workbench instances includes the following:

- **Gemini CLI Update**: Minor updates to Gemini CLI's dependencies.
Feature **Workbench Image Release 26.03**

Released the new 26.03 image version under the following image family `workbench-instances-2603`. This major update introduces:

- **OS Upgrade**: Migration to Debian 12 (Bookworm) for improved security and performance.
- **Python Upgrade**: Python 3.12 as the default version.
- **New Versioning Scheme** : Transitioned to a date-based versioning format (e.g., v26.03) to provide better historical context for image updates. For more information on new versioning schema see the [latest documentation](https://cloud.google.com/vertex-ai/docs/workbench/instances/manage-image-versions)
- **Micromamba CLI**: Installed by default
- **JupyterLab Environments**: Supports only JupyterLab 4.
Change

### M140 Release

Change **Gemini CLI Update**: Minor updates to Gemini CLI's dependencies.

## March 25, 2026

**Generative AI on Vertex AI** v1 Feature **Lyria 3**

Lyria is available in [public
preview](https://cloud.google.com/products#product-launch-stages). You can use
`lyria-3-pro-preview` to generate 184 seconds of audio, or
`lyria-3-clip-preview` to generate 30 seconds of audio.

For more information, see the following:

- [Lyria 3 Pro
  Preview](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/lyria/lyria-3#lyria-3-pro-preview)

- [Lyria 3 Clip Preview](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/lyria/lyria-3#lyria-3-clip-preview)

## March 24, 2026

**Generative AI on Vertex AI** v1 Deprecated **Imagen generation GA endpoints deprecation**

The following table describes image generation endpoints that are deprecated and
their replacements. We recommend updating your model endpoints before June 30,
2026, to avoid service disruption.

| Discontinued endpoints | Recommended endpoint migration |
|---|---|
| `imagegeneration@002` | `gemini-2.5-flash-image` |
| `imagegeneration@003` | `gemini-2.5-flash-image` |
| `imagegeneration@004` | `gemini-2.5-flash-image` |
| `imagegeneration@005` | `gemini-2.5-flash-image` |
| `imagegeneration@006` | `gemini-2.5-flash-image` |
| `imagetext@001` | `gemini-2.5-flash-image` |
| `imagen-3.0-capability-001` | `gemini-2.5-flash-image` |
| `imagen-3.0-capability-002` | `gemini-2.5-flash-image` |
| `imagen-3.0-fast-generate-001` | `gemini-2.5-flash-image` |
| `imagen-3.0-generate-001` | `gemini-2.5-flash-image` |
| `imagen-3.0-generate-002` | `gemini-2.5-flash-image` |
| `imagen-4.0-fast-generate-001` | `gemini-2.5-flash-image` |
| `imagen-4.0-generate-001` | `gemini-2.5-flash-image` |
| `imagen-4.0-ultra-generate-001` | `gemini-2.5-flash-image` |

Deprecated **Video generation GA endpoints deprecation**

The following table describes video generation endpoints that are deprecated and
their replacements. We recommend updating your model endpoints before June 30,
2026, to avoid service disruption.

| Discontinued endpoints | Recommended endpoint migration |
|---|---|
| `veo-3.0-generate-001` | `veo-3.1-generate-001` |
| `veo-3.0-fast-generate-001` | `veo-3.1-fast-generate-001` |
| `veo-2.0-generate-001` | `veo-3.1-generate-001` |

## March 16, 2026

**Vertex AI** v1 Deprecated Vertex Explainable AI is deprecated. For details, see [Vertex AI deprecations](https://docs.cloud.google.com/vertex-ai/docs/deprecations.md.txt).

## March 12, 2026

**Generative AI on Vertex AI** v1 Feature **Partner model evaluations**

The Gen AI evaluation service supports evaluating partner models, such as Anthropic
and Llama models. For more information, see
[Perform evaluation using the console](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/evaluation-genai-console#evaluate_partner_models).

## March 05, 2026

**Vertex AI** v1 Announcement **Vector Search 2.0 is generally available (GA)**

Vector Search 2.0 is generally available ([GA](https://cloud.google.com/products/#product-launch-stages)).

Vector Search 2.0 is a retrieval engine designed to streamline AI development
and serve as the knowledge core for AI applications.

Available features:

- [Collections](https://docs.cloud.google.com/vertex-ai/docs/vector-search-2/collections/collections) unifying data and vectors.
- Auto-embeddings for populating vector fields automatically.
- Hybrid search and ranking combining vector, full-text, and built-in semantic re-ranking in a single parallel query.
- Private Service Connect (PSC), Private Google Accesss (PGA), and VPC Service Controls support.

For more information, see the [Vector Search 2.0 documentation](https://docs.cloud.google.com/vertex-ai/docs/vector-search-2/overview).

## March 03, 2026

**Generative AI on Vertex AI** v1 Deprecated **Video generation preview endpoints deprecation**

The following table describes video generation endpoints that are deprecated and
their replacements. We recommend updating your model endpoints before April 2,
2026, to avoid service disruption.

| Discontinued endpoints | Recommended endpoint migration |
|---|---|
| `veo-3.0-generate-preview` | `veo-3.0-generate-001` |
| `veo-3.0-fast-generate-preview` | `veo-3.0-fast-generate-preview` |
| `veo-2.0-generate-preview` | `veo-2.0-generate-001` |
| `veo-2.0-generate-exp` | `veo-2.0-generate-001` |
| `veo-001-preview-0815` | `veo-2.0-generate-001` |
| `veo-001-preview` | `veo-2.0-generate-001` |
| `veo-3.1-generate-preview` | `veo-3.1-generate-001` |
| `veo-3.1-fast-generate-preview` | `veo-3.1-fast-generate-001` |

Feature **Gemini 3.1 Flash-Lite**

Gemini 3.1 Flash-Lite (`gemini-3.1-flash-lite`) is
available in [public preview](https://cloud.google.com/products#product-launch-stages).
This release is our most cost-efficient Gemini model and is
optimized for low latency use cases for high-volume, cost-sensitive LLM traffic.

For more information, see
[Gemini 3.1 Flash-Lite](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/3-1-flash-lite).

## February 26, 2026

**Generative AI on Vertex AI** v1 Feature **Gemini 3.1 Flash Image**

Gemini 3.1 Flash Image (`gemini-3.1-flash-image`) is available
in [public preview](https://cloud.google.com/products#product-launch-stages).
This release enables high-quality image generation with improved pricing and
latency. We recommend using Gemini 3.1 Flash Image when generating
images.

For more information, see
[Gemini 3.1 Flash Image](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/3-1-flash-image).

## February 23, 2026

**Generative AI on Vertex AI** v1 Deprecated **Anthropic's Claude 3 Haiku**

Anthropic's Claude 3 Haiku is deprecated as of February 23, 2026 and will be
shut down on August 23, 2026. For more information, see
[Partner model deprecations](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/deprecations/partner-models#haiku-3).

## February 20, 2026

**Agent Platform Workbench** v2 Fixed Fixed a vulnerability in Vertex AI Workbench for users utilizing
managed credentials.

## February 19, 2026

**Generative AI on Vertex AI** v1 Feature **Gemini 3.1 Pro Preview**

[Gemini 3.1 Pro](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/3-1-pro)
is available in preview in Model Garden. Gemini 3.1 Pro is
our most advanced reasoning Gemini model, capable of solving complex
problems from different information sources, including text, audio, images,
video, PDFs, and even entire code repositories with its 1M token context
window.

## February 17, 2026

**Generative AI on Vertex AI** v1 Feature **Anthropic's Claude Sonnet 4.6**

[Claude Sonnet 4.6](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/claude/sonnet-4-6)
is available in Model Garden.
Deprecated **Image generation preview endpoints deprecation**

The following table describes image generation endpoints that are deprecated and
their replacements. We recommend updating your model endpoints before March 19,
2026, to avoid service disruption.

| Discontinued endpoints | Recommended endpoint migration |
|---|---|
| `gemini-2.0-flash-image-generation-preview` | `gemini-2.5-flash-image` |
| `gemini-2.5-flash-image-generation-preview` | `imagen-4.0-generate-001` or `gemini-2.5-flash-image` |
| `imagen-4.0-generate-preview-05-20` | `imagen-4.0-generate-001` or `gemini-2.5-flash-image` |
| `imagen-4.0-generate-preview-06-06` | `imagen-4.0-generate-001` or `gemini-2.5-flash-image` |
| `imagen-4.0-ultra-generate-preview-06-06` | `imagen-4.0-generate-001` or `gemini-2.5-flash-image` |
| `imagen-4.0-fast-generate-preview-05-20` | `imagen-4.0-generate-001` or `gemini-2.5-flash-image` |
| `imagen-product-recontext-preview-06-30` | `gemini-2.5-flash-image` |
| `imagen-2.0-edit-preview-0627` | `gemini-2.5-flash-image` |
| `virtual-try-on-preview-08-04` | `virtual-try-on-001` |
| `imagen-4.0-ingredients-preview` | `gemini-2.5-flash-image` |

**Vertex AI** v1 Deprecated Vertex AI Feature Store Optimized online serving is deprecated. For details, see [Vertex AI deprecations](https://docs.cloud.google.com/vertex-ai/docs/deprecations.md.txt).
Deprecated Vertex AI Feature Store (Legacy/V1) is deprecated. For details, see [Vertex AI deprecations](https://docs.cloud.google.com/vertex-ai/docs/deprecations.md.txt).

## February 10, 2026

**Generative AI on Vertex AI** v1 Feature **GLM 5** is available as an experimental launch in Model Garden. This model
is targeting complex systems engineering and long-horizon agentic tasks.
GLM 5 is available as a managed API in Model Garden. To learn more, see
[GLM 5](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/zai-org/glm-5).

## February 04, 2026

**Generative AI on Vertex AI** v1 Feature **Anthropic's Claude Opus 4.6**

[Claude Opus 4.6](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/claude/opus-4-6)
is available in Model Garden.

## January 29, 2026

**Agent Platform Workbench** v2 Feature **M139 release**

The M139 release of Vertex AI Workbench instances includes the following:

- Removed a startup script dependency on `gs://dl-platform-public-configs`.
Change

### M139 Release

Change Removed a startup script dependency on `gs://dl-platform-public-configs`.

## January 23, 2026

**Generative AI on Vertex AI** v1 Feature **Virtual Try-On**

Virtual Try-On is now [generally available
(GA)](https://cloud.google.com/products/#product-launch-stages).
The new endpoint, `virtual-try-on-001`,
replaces the previous endpoint, `virtual-try-on-preview-08-04`. We
recommend changing to the new endpoint as soon as possible.

For more information, see
[Generate Virtual Try-On Images](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/generate-virtual-try-on-images.md.txt).

## January 22, 2026

**Generative AI on Vertex AI** v1 Announcement Codestral (25.01) and Mistral Large (24.11) are retired as of January 23, 2026.

## January 21, 2026

**Colab Enterprise** Feature **BigFrames, BigQuery ML, and Managed Service for Apache Spark**

The Data Science Agent supports BigFrames, BigQuery ML, and
Managed Service for Apache Spark. Previously, these capabilities were supported
only when using Colab Enterprise notebooks in BigQuery.

## January 20, 2026

**Generative AI on Vertex AI** v1 Feature **GLM 4.7** GA is now available in Model Garden. This model
is designed for core or vibe coding, tool use, and complex reasoning.
GLM 4.7 is available as a managed API in Model Garden. To learn more, see
[GLM 4.7](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/zai-org/glm-47).

## January 16, 2026

**Agent Platform Workbench** v2 Feature [Preview](https://cloud.google.com/products#product-launch-stages): The Gemini
CLI is available in Vertex AI Workbench instances. The Gemini CLI is an open
source AI agent that provides access to Gemini directly in a terminal. You can
use the Gemini CLI to do tasks like the following:

- Create a new notebook.
- Run notebook cells.
- Write and edit a notebook's code and text cells.
- Explain code and technical concepts.
- Interact with a Vertex AI Workbench instance's local file system, including performing complex file operations that span multiple files based on a single, high-level instruction.
- Run basic shell commands.
- Run commands to interact with other Google Cloud services, such as Vertex AI and BigQuery.

To get started, see
[Use the Gemini CLI](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/gemini-cli).

## January 14, 2026

**Agent Platform Workbench** v2 Feature **M138 release**

The M138 release of Vertex AI Workbench instances includes the following:

- Fixed an issue where daemon processes launched by post-startup scripts, such as gcsfuse mounts, did not persist after the script finished executing.
Change

### M138 Release

Fixed Fixed an issue where daemon processes launched by post-startup scripts, such as
gcsfuse mounts, did not persist after the script finished executing.

## January 13, 2026

**Generative AI on Vertex AI** v1 Feature **Veo 3.1 reference-to-video update**

Veo 3.1 Preview models now support the following features:

- 9:16 aspect ratio for reference-to-video.
- Upsampling for videos generated at 1080p and 4k resolutions.

For more information, see the following:

- [Generate Veo videos from reference images](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/use-reference-images-to-guide-video-generation)
- [Veo on Vertex AI video generation API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/veo-video-generation)

## January 06, 2026

**Generative AI on Vertex AI** v1 Feature **GLM 4.7** is available as an experimental launch in Model Garden. This model
is designed for core or vibe coding, tool use, and complex reasoning.
GLM 4.7 is available as a managed API in Model Garden. To learn more, see
[GLM 4.7](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/zai-org/glm-47).

## January 05, 2026

**Generative AI on Vertex AI** v1 Deprecated **Anthropic's Claude 3.5 Haiku**

Anthropic's Claude 3.5 Haiku is deprecated as of January 5, 2026 and will be
shut down on July 5, 2026. For more information, see
[Partner model deprecations](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/deprecations/partner-models#haiku-3-5).

## December 18, 2025

**Generative AI on Vertex AI** v1 Feature The following models are available through Model Garden:

- [FunctionGemma](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/functiongemma)
- [T5Gemma 2](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/t5gemma)
Feature **Save and share prompts in Vertex AI Studio** : The prompt sharing feature no longer needs to be enabled. You can share prompts without asking your administrator to first enable the prompt sharing feature. For more information, see [Save and share prompts](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/prompt-sharing).

## December 17, 2025

**Colab Enterprise** Feature **Post-startup scripts**

[Generally available (GA)](https://cloud.google.com/products#product-launch-stages):
You can use a post-startup script to perform tasks after the startup process
of your Colab Enterprise runtime. For example, you can use
a post-startup script to install specific packages or make specific
changes to your runtime's VM. For more information, see
[Use a post-startup script](https://docs.cloud.google.com/colab/docs/post-startup-script.md.txt).
**Generative AI on Vertex AI** v1 Feature [Cloud API Registry](https://docs.cloud.google.com/api-registry/docs/overview.md.txt) is available
in the Google Cloud console in
[Preview](https://cloud.google.com/products#product-launch-stages). Use
Cloud API Registry in the Google Cloud console to view and manage the MCP
servers and tools your agent has access to.
Feature **Gemini 3 Flash**

Gemini 3 Flash is now available in public preview. This model is designed to
tackle the most challenging agentic problems with strong coding and
state-of-the-art reasoning capabilities, and is our best model for complex
multimodal understanding.

For more information, see [Gemini 3
Flash](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/3-flash).

## December 16, 2025

**Generative AI on Vertex AI** v1 Change Updated pricing for **Vertex AI Agent Engine**:

- Pricing for Vertex AI Agent Engine Runtime was lowered.

- On **January 28, 2026**, Sessions, Memory Bank, and Code Execution will begin
  charging for usage.

For more information, see [Pricing](https://cloud.google.com/vertex-ai/pricing#vertex-ai-agent-engine).
Announcement **Vertex AI Agent Engine**

Vertex AI Agent Engine [Sessions](https://docs.cloud.google.com/agent-builder/agent-engine/sessions/overview)
and [Memory Bank](https://docs.cloud.google.com/agent-builder/agent-engine/memory-bank/overview) are now
[Generally Available](https://cloud.google.com/products#product-launch-stages).
Change **Vertex AI Agent Engine**

Vertex AI Agent Engine is now available in the following regions:

- `europe-west6` (Zurich)
- `europe-west8` (Milan)
- `asia-east2` (Hong Kong)
- `asia-northeast3` (Seoul)
- `asia-southeast2` (Jakarta)
- `northamerica-northeast2` (Toronto)
- `southamerica-east1` (São Paulo)

For more information, see [Vertex AI Agent Builder locations](https://docs.cloud.google.com/agent-builder/locations).
Change **Virtual Try-On**

Our Virtual Try-On model, `virtual-try-on-preview-08-04`, is
improved. Latency is significantly reduced and quality is improved for shoes,
body shape preservation, and product fidelity.

## December 12, 2025

**Agent Platform Workbench** v2 Feature
- The following features are now [generally available (GA)](https://cloud.google.com/products#product-launch-stages):

  - [Create new Vertex AI Workbench instances](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/create#create) with JupyterLab 4+.

  - [Create an instance with managed user credential access.](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/create-euc-instance)

- Using post-startup scripts for new instances with user credential access is now available in private GA. For information about access to this release, see the [access request page](https://docs.google.com/forms/d/e/1FAIpQLSdzNZUFsRRbB0KaMCObGUYzIo6lK3X2tO7JZOm_DZrUhBYw5Q/viewform). For more information, see [Create instances with post-startup scripts](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/create-euc-instance#post-startup-scripts).

**Generative AI on Vertex AI** v1 Feature **Gemini 2.5 Flash with Gemini Live API Native Audio**

[Gemini 2.5 Flash with Gemini Live API Native Audio](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash-live-api) (`gemini-live-2.5-flash-native-audio`) is Generally Available (GA).
This model features cutting-edge native audio functionality for
[Gemini Live API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/live-api), including enhanced voice quality and adaptability, Proactive Audio, and Affective Dialog.

## December 10, 2025

**Agent Platform Workbench** v2 Feature **M137 release**

The M137 release of Vertex AI Workbench instances includes the following:

- Patched an issue where `sudo` prompts for a password when a custom Jupyter user name is specified.
Change

### M137 Release

Fixed Patched an issue where `sudo` prompts for a password when a custom Jupyter user
name is specified.
**Generative AI on Vertex AI** v1 Feature **DeepSeek-V3.2** is available in Model Garden.
DeepSeek-V3.2 is a state-of-the-art large language model from
DeepSeek.
DeepSeek-V3.2 is available as a managed API in Model Garden. To learn more, see
[DeepSeek-V3.2](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/deepseek/deepseek-v32).

## December 09, 2025

**Agent Platform Workbench** v2 Feature [Generally available (GA)](https://cloud.google.com/products#product-launch-stages):
Encrypt your data-in-use by using
[Confidential Computing](https://docs.cloud.google.com/confidential-computing/docs/confidential-computing-overview.md.txt).
You can enable the Confidential VM service when you create
a Vertex AI Workbench instance. To get started, see [Create an instance with
Confidential Computing](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/create-confidential-computing).
**Generative AI on Vertex AI** v1 Feature The following models are available through Model Garden:

- [Ministral 3](https://console.cloud.google.com/vertex-ai/publishers/mistralai/model-garden/ministral-3)
- [Mistral Large 3](https://console.cloud.google.com/vertex-ai/publishers/mistralai/model-garden/mistral-large-3)

## December 08, 2025

**Generative AI on Vertex AI** v1 Feature **Veo 3.1 video extension**

Veo 3.1 supports video extension in Preview.

For more information, see the following:

- [Extend Veo
  videos](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/extend-a-veo-video)
- [Veo video generation API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/veo-video-generation)

## December 02, 2025

**Generative AI on Vertex AI** v1 Feature The following models are available through Model Garden:

- [DeepSeek-V3.2](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-v3-2;publisherModelVersion=deepseek-v3-2)
- [DeepSeek-V3.2-Speciale](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-v3-2;publisherModelVersion=deepseek-v3-2-speciale)
Feature The Vertex AI Model Garden model co-hosting vLLM container is available to use with [this sample notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_model_cohost.ipynb). You can use this container to serve multiple replicas of a model and serve multiple models with dynamic loading and unloading. This allows you to maximize resource utilization and serving efficiency, and flexibly adjust the models to serve.

## November 24, 2025

**Generative AI on Vertex AI** v1 Feature **Anthropic's Claude Opus 4.5**

[Claude Opus 4.5](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/claude/opus-4-5)
is available in Model Garden.

## November 21, 2025

**Agent Platform Workbench** v2 Feature **M136 release**

The M136 release of Vertex AI Workbench instances includes the following:

- Patched an issue where image outputs aren't displayed properly.
Change

### M136 Release

Fixed Patched an issue where image outputs aren't displayed properly.

## November 17, 2025

**Agent Platform Workbench** v2 Feature **M135 release**

The M135 release of Vertex AI Workbench instances includes the following:

- Patched an issue where user-triggered OS shutdowns aren't reported to the Notebooks API.
Change

### M135 Release

Fixed Patched an issue where user-triggered OS shutdowns aren't reported to the
Notebooks API.
**Generative AI on Vertex AI** v1 Feature **Veo video generation**

Veo 3.1 is Generally Available, and introduces the following models:

- [Veo
  3.1](https://cloud.google.com/vertex-ai/generative-ai/docs/models/veo/3-1-generate#3.1-generate-001)
- [Veo 3.1
  Fast](https://cloud.google.com/vertex-ai/generative-ai/docs/models/veo/3-1-generate#3.1-fast-generate-001)

For more information, see the following:

- [Generate videos with Veo on Vertex AI](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/overview)
- [Generate Veo videos from text prompts](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-text)
- [Generate Veo videos from an image](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-an-image)
- [Generate Veo videos using first and last frames](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-first-and-last-frames)
- [Veo video generation API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/veo-video-generation)
Announcement **LearnLM in Gemini**

The LearnLM model is no longer a separate offering or listing on AI Studio as
[LearnLM capabilities have been integrated into the latest Gemini models (starting with Gemini 2.5)](https://blog.google/outreach-initiatives/education/google-gemini-learnlm-update/).

Built in collaboration with experts in education,
[LearnLM](https://cloud.google.com/solutions/learnlm) represents our
capabilities fine-tuned for learning informed by rigorous research. These
advancements and improvements are available directly in Gemini, enhancing
educational experiences and applications.

Pre-existing learnlm-2.0-flash-experimental projects will not remain functional
past December 3, 2025 unless an alternative model is manually selected---we
encourage developers to switch to the latest Gemini models and optimize their
prompts by reviewing our
[LearnLM Partner Prompt Guide](https://services.google.com/fh/files/misc/learnlm_prompt_guide.pdf).

## November 13, 2025

**Generative AI on Vertex AI** v1 Feature **Kimi K2 Thinking** is available in Model Garden. This model is
a thinking model that excels at complex problem-solving and deep reasoning.
Kimi K2 Thinking is available as a managed API in Model Garden. To learn more, see
[Kimi K2 Thinking](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/kimi/kimi-k2-thinking).
Feature **Updated Prompt Caching for Anthropic Claude Models**

Prompt caching for Anthropic Claude models now supports a one-hour Time To Live (TTL).

For more information, see [Prompt caching](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/claude/prompt-caching).

## November 11, 2025

**Generative AI on Vertex AI** v1 Deprecated **Anthropic's Claude 3.7 Sonnet**

Anthropic's Claude 3.7 Sonnet is deprecated as of November 11, 2025 and will be
shut down on May 11, 2026. For more information, see
[Partner model deprecations](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/deprecations/partner-models#sonnet-3-7).

## November 10, 2025

**Colab Enterprise** Feature The default latest Python version is now 3.12. See
[Python versions](https://docs.cloud.google.com/colab/docs/runtimes.md.txt#python).

## November 07, 2025

**Generative AI on Vertex AI** v1 Feature **Vertex AI Agent Engine**

The following features are now available in [Preview](https://cloud.google.com/products#product-launch-stages):

- Configure, manage, and view [observability
  features](https://docs.cloud.google.com/agent-builder/agent-engine/manage/overview#configure_telemetry_for_your_deployed_agent)
  such as sessions, traces, logs, and events for your agent in the Google Cloud console.

- [Use the
  playground](https://docs.cloud.google.com/agent-builder/agent-engine/use/overview#use-deployed-agents-console)
  to test and interact with your agent in the Google Cloud console.

- [Evaluate your agents](https://docs.cloud.google.com/agent-builder/agent-engine/evaluate) using the
  Gen AI evaluation service's GenAI Client in Vertex AI SDK.

- [Create and manage memory
  revisions](https://docs.cloud.google.com/agent-builder/agent-engine/memory-bank/revisions) for Memory Bank.

- Use Identity Access Management (IAM) to create an [agent
  identity](https://docs.cloud.google.com/agent-builder/agent-engine/agent-identity) to manage access and
  authentication when using agents on Vertex AI Agent Engine Runtime.

The following features are now available in [GA](https://cloud.google.com/products#product-launch-stages):

- [Express mode support](https://docs.cloud.google.com/agent-builder/agent-engine/overview#express-mode) for
  Vertex AI Agent Engine Runtime.

- Use the new free tier with Vertex AI Agent Engine Runtime. For more
  information, see
  [Pricing](https://cloud.google.com/vertex-ai/pricing#vertex-ai-agent-engine).

## November 04, 2025

**Generative AI on Vertex AI** v1 Feature **MiniMax M2** is available in Model Garden. This model is
is built for end-to-end development workflows and has strong capabilities
in planning and executing complex tool-calling tasks. The model is
optimized to provide a balance of performance, cost, and inference speed.
MiniMax M2 is available as a managed API in Model Garden. To learn more, see
[MiniMax M2](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/minimax/minimax-m2).

## October 23, 2025

**Generative AI on Vertex AI** v1 Feature The following models are available through Model Garden:

- [DeepSeek-OCR](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-ocr)
- [Qwen3-VL](https://console.cloud.google.com/vertex-ai/publishers/qwen/model-garden/qwen3-vl)
- [Earth AI](https://console.cloud.google.com/vertex-ai/model-garden/google/earth-ai)

## October 21, 2025

**Generative AI on Vertex AI** v1 Security On September 23, 2025, we discovered a technical issue in
the Vertex AI API that resulted in a limited amount of responses
being misrouted between recipients for certain third-party models
when using streaming requests. This issue is now resolved.
Google models, e.g. Gemini, were not impacted.

Some internal proxies did not properly handle HTTP requests that
have an `Expect: 100-continue` header, resulting in
a desynchronization in a streaming response connection, where
a response intended for one request was instead delivered as
the response for a subsequent request.

For more information, see
[Security bulletins](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/security-bulletins#gcp-2025-059).

## October 20, 2025

**Colab Enterprise** Feature **Visualization cells**

[Preview](https://cloud.google.com/products#product-launch-stages): You can
use visualization cells to generate interactive and editable visualizations
from within a Colab Enterprise notebook. You can configure the
chart type, aggregation, colors, labels, and other aspects of the
visualization to help you explore data and discover insights. For more
information, see [Use visualization cells](https://docs.cloud.google.com/colab/docs/visualization-cells.md.txt).

## October 16, 2025

**Generative AI on Vertex AI** v1 Feature **Mistral's Codestral 2**

You can use Mistral's
[Codestral 2](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/mistral/codestral-2)
in Model Garden.
Feature **vLLM TPU**

[vLLM TPU](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/open-models/vllm/use-vllm-tpu), a
highly-efficient serving framework for large language models (LLM) that's
optimized for [Cloud TPU](https://docs.cloud.google.com/tpu) hardware, is available through
Model Garden.

## October 15, 2025

**Generative AI on Vertex AI** v1 Feature **Anthropic's Claude Haiku 4.5**

You can use Anthropic's
[Claude Haiku 4.5](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/claude/haiku-4-5)
in Model Garden.
Feature **Veo video generation**

Veo 3.1 is available in Preview, and introduces the following models:

- [Veo
  3.1 preview](https://cloud.google.com/vertex-ai/generative-ai/docs/models/veo/3-1-generate-preview)
- [Veo 3.1
  Fast preview](https://cloud.google.com/vertex-ai/generative-ai/docs/models/veo/3-1-fast-generate-preview).

For more information, see the following:

- [Generate videos with Veo on Vertex AI](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/overview)
- [Generate Veo videos from text prompts](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-text)
- [Generate Veo videos from an image](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-an-image)
- [Generate Veo videos using first and last frames](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-first-and-last-frames)
- [Direct Veo using a reference image](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/use-reference-images-to-guide-video-generation)
- [Veo video generation API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/veo-video-generation)
Feature **Veo video generation**

Veo 2 supports adding and removing objects from videos in Preview.

For more information about Veo 2, see [Veo 2
Preview](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/veo/2-0-generate-preview)

For more information about adding and removing objects, see the following:

- [Insert objects into Veo videos](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/insert-objects-into-videos)
- [Remove objects from Veo videos](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/remove-objects-from-videos)
- [Veo video generation API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/veo-video-generation)

## October 14, 2025

**Colab Enterprise** Feature **SQL cells**

[Preview](https://cloud.google.com/products#product-launch-stages): You can
use SQL cells to write, edit, and run SQL queries directly from your
Colab Enterprise notebooks. For more information, see
[Use SQL cells](https://docs.cloud.google.com/colab/docs/sql-cells.md.txt).
**Generative AI on Vertex AI** v1 Deprecated **Imagen subject and style fine-tuning**

Imagen subject model and style model tuning will be removed on
December 31, 2025. We recommend that you use [Gemini 2.5 Flash Image](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash-image), which supports most use cases that require
fine-tuning. For more information, see [Edit images with
Gemini](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/image-editing).
Deprecated **Imagen 4 preview models**

The following Imagen 4 preview models will be removed on
November 30, 2025:
`imagen-4.0-generate-preview-06-06`,
`imagen-4.0-ultra-generate-preview-06-06`, and
`imagen-4.0-fast-generate-preview-06-06`. To avoid service
disruption, migrate all workflows that use Imagen 4 preview models before
November 30, 2025, 2025, to the following Imagen 4 Generally
Available models: `imagen-4.0-generate-001`,
`imagen-4.0-ultra-generate-001`,
`imagen-4.0-fast-generate-001`.

## October 09, 2025

**Agent Platform Workbench** v2 Feature **M134 release**

The M134 release of Vertex AI Workbench instances includes the following:

- Patched a regression with custom notebook metrics reporting (for example, `jupyterlab_kernels` and `docker_status metrics`).
- Updated the Dataproc JupyterLab plugin (`dataproc-jupyter-plugin`) to version 0.1.92.
- When using Google Cloud CLI commands, the `project` and `region` properties are preset.
Change

### M134 Release

Fixed Patched a regression with custom notebook metrics reporting (for example,
`jupyterlab_kernels` and `docker_status metrics`).
Change Updated the Dataproc JupyterLab plugin (`dataproc-jupyter-plugin`) to version
0.1.92.
Change When using Google Cloud CLI commands, the `project` and `region` properties are
preset.
**Generative AI on Vertex AI** v1 Change **Imagen**

Imagen's virtual try-on model, `virtual-try-on-preview-08-04`
was updated on September 30, 2025, to more accurately preserve the person's
body shape and preserve the garment's identity.

## October 07, 2025

**Colab Enterprise** Feature **Post-startup scripts**

[Preview](https://cloud.google.com/products#product-launch-stages): You can
use a post-startup script to perform tasks after the startup process
of your Colab Enterprise runtime. For example, you can use
a post-startup script to install specific packages or make specific
changes to your runtime's VM. For more information, see
[Use a post-startup script](https://docs.cloud.google.com/colab/docs/post-startup-script.md.txt).
**Generative AI on Vertex AI** v1 Feature The following Qwen models are available in [Model Garden](https://console.cloud.google.com/vertex-ai/publishers/qwen/model-garden/qwen-image):

- Qwen-Image
- Qwen-Image-Edit
- Qwen-Image-Edit-2509
v1beta1 Announcement The [Computer Use tool](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/computer-use) (`gemini-2.5-computer-use-preview-10-2025`) is now available in Preview. The Computer Use model and tool lets you enable your applications to interact with and automate tasks in the browser. With the Computer Use model and tool, you can build agents that can:

- Automate repetitive data entry or form filling on websites.

- Navigate websites to gather information.

- Assist users by performing sequences of actions in web applications.

## October 06, 2025

**Generative AI on Vertex AI** v1 Change **Updated pricing for Vertex AI Agent Engine** : Starting on **November 6, 2025**, Vertex AI Agent Engine Runtime will start charging for runtime usage for the following regions:

- `asia-southeast1` (Singapore)
- `australia-southeast2` (Melbourne)
- `europe-west2` (London)
- `europe-west3` (Frankfurt)
- `europe-west4` (Netherlands)

For more details, see [Pricing](https://cloud.google.com/vertex-ai/pricing#agent_engine) for Vertex AI Agent Engine.
Feature **Access Transparency for Vertex AI Agent Engine** : Access Transparency is now available for Vertex AI Agent Engine. For more information, see the overview for [Enterprise security](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/overview#enterprise_security).

## October 03, 2025

**Generative AI on Vertex AI** v1 Feature **Prompt management**

Vertex AI offers tooling to help manage prompts and prompt versions. In addition to the prompt management capabilities in Vertex AI Studio, prompts can be stored and versioned using the Vertex AI SDK.

For more information, see the [Prompt management](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/prompt-classes) API reference.

## October 02, 2025

**Generative AI on Vertex AI** v1 Announcement **[Gemini 2.5 Flash Image](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash#image-generation)** (`gemini-2.5-flash-image`) is now generally available. This GA release adds support for aspect ratio controls, image-only response modality, regional endpoints, [support for batch predictions](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/batch-prediction-gemini), [image generation from multiple reference images](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/image-generation#image-generation), and [improved multi-turn image editing](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/image-editing#multi-turn-editing).

See [Gemini 2.5 Flash Image](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash#image) for more information.
Feature **Google Gen AI SDK in C# Preview**

[Preview](https://cloud.google.com/products#product-launch-stages): The Google Gen AI SDK is available in C#. See [googleapis/dotnet-genai](https://github.com/googleapis/dotnet-genai).

This release includes support for `GenerateContentAsync`, `GenerateContentStreamAsync`, `GenerateImagesAsync`, and three Live APIs, which includes `SendClientContentAsync`, `SendRealtimeInputAsync`, and `SendToolResponseAsync`.

## October 01, 2025

**Agent Platform Workbench** v2 Feature [Generally available (GA)](https://cloud.google.com/products/#product-launch-stages):
You can use [Workforce Identity Federation](https://docs.cloud.google.com/iam/docs/workforce-identity-federation.md.txt)
with Vertex AI Workbench instances. Workforce Identity Federation lets
you create and manage Vertex AI Workbench instances with credentials
provided by an external identity provider (IdP). For more information, see
[Create an instance with third party credentials](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/create-third-party-instance).

## September 30, 2025

**Generative AI on Vertex AI** v1 Feature [DeepSeek-V3.2-Exp](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-v3-2) is available through Model Garden.
**Vertex AI** v1 Feature [DeepSeek-V3.2-Exp](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-v3-2) is available through Model Garden.

## September 25, 2025

**Generative AI on Vertex AI** v1 Announcement New preview models for [**Gemini 2.5 Flash**](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash#2.5-flash-preview) and [**2.5 Flash-Lite**](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash-lite#2.5-flash-lite-preview) are now available. These models are available at the following versioned endpoints:

- `gemini-2.5-flash-preview-09-2025`
- `gemini-2.5-flash-lite-preview-09-2025`

## September 24, 2025

**Generative AI on Vertex AI** v1 Deprecated Access to Gemini's 1.5 models has been discontinued. For more information, see our [Model versions](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/model-versions) page.

## September 23, 2025

**Generative AI on Vertex AI** v1 Announcement **Gemini 2.5 Flash with Live API Native Audio Preview**

[**Gemini 2.5 Flash with Live API Native Audio**](https://docs.cloud.google.com/release-notes/cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash) (`gemini-live-2.5-flash-preview-native-audio-09-2025`) is available in [Preview](https://cloud.google.com/products#product-launch-stages). A single, unified model processes audio input and generates audio output directly, eliminating separate text-to-speech/speech-to-text conversions. This results in-low latency, high-quality, and incredibly human-like conversations. New features and capabilities include:

- **Improved Barge-in:** Interrupt Gemini more naturally and reliably, even in loud and noisy environments.

- **Robust Function Calling:** We've improved the triggering rate, allowing Gemini to successfully execute the functions you define with greater precision.

- **Accurate Transcription:** The accuracy of audio-to-text transcription has been significantly enhanced.

- **Seamless Multilingual Support:** Speak to Gemini in multiple languages, and it will effortlessly switch between them without any pre-configuration. Language is no longer a barrier!

- **Enhanced Audio Quality:** Experience a dramatically improved audio quality that truly feels like speaking with a person.

- **Proactive Audio:** Define Gemini's expertise and set conditions for when it should respond. Gemini can act as a "silent listener," only chiming in when the conversation touches upon its designated area of expertise.

- **Affective Dialog:** Gemini can adapt and adjust its generated voice to match the emotional tone of the speaker, creating more empathetic and natural interactions.

[Watch our comprehensive demo](https://youtu.be/ZsB33Tr-P3c) to see these features in action, including seamless language switching, expert mode, emotionally aware responses, memory recall, and interactive screen sharing for engineering tasks -- all demonstrated directly within Vertex AI Studio without writing a single line of code!

## September 22, 2025

**Generative AI on Vertex AI** v1 Feature [DeepSeek-V3.1-Terminus](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-v3-1;publisherModelVersion=deepseek-v3-1-terminus) is available through Model Garden.
**Vertex AI** v1 Feature [DeepSeek-V3.1-Terminus](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-v3-1;publisherModelVersion=deepseek-v3-1-terminus) is available through Model Garden.

## September 18, 2025

**Generative AI on Vertex AI** v1 Change **Grounding with Google Maps**

Grounding with Google Maps has implemented the following changes:

- Removed the following fields from the API response:
  - `grounding_chunk.maps.text`
  - `grounding_chunk.maps.place_answer_sources.review_snippets.author_attribution`
  - `grounding_chunk.maps.place_answer_sources.flag_content_uri`
  - `grounding_chunk.maps.place_answer_sources.review_snippets.flag_content_uri`
- The widget context token is only returned when the optional `widget_token_enable` input flag is set.

To learn more, see [Grounding with Google Maps](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/grounding/grounding-with-google-maps).

## September 17, 2025

**Agent Platform Workbench** v2 Feature **M133 release**

The M133 release of Vertex AI Workbench instances includes the following:

- Patched an incompatibility with the Dataproc JupyterLab plugin (`dataproc-jupyter-plugin`) and instances with end-user credentials enabled.
Change

### M133 Release

Fixed Patched an incompatibility with the Dataproc JupyterLab plugin
(`dataproc-jupyter-plugin`) and instances with end-user credentials enabled.

## September 15, 2025

**Generative AI on Vertex AI** v1 Change **Imagen**

We improved Imagen's virtual try-on model, `virtual-try-on-preview-08-04`, so that it is better at preserving the person's body shape and preserving the garment product's identity.
**Vertex AI** v1 Feature **Vector Search indexes optimized for storage (Preview)**

Storage-optimized Vector Search offers a cost-effective solution for searching massive datasets. This new tier is ideal for large-scale RAG and semantic search applications, and simplifies development with a new auto-tuning feature that eliminates the need to manage the underlying index configuration. This feature is available in Preview.

For more information, see [Storage-optimized Vector Search](https://docs.cloud.google.com/vertex-ai/docs/vector-search/storage-optimized-vector-search).

## September 10, 2025

**Generative AI on Vertex AI** v1 Breaking **Vertex AI Agent Engine**

In version `v1.112.0` of the Vertex AI SDK for Python, the `agent_engines` module has been refactored to a [client-based design](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/overview#migration). For information about updating your existing code to the new design, see the [Migration guide](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/deprecations/agent-engine-migration).
Feature **Vertex AI Agent Engine**

Agent Engine now supports the following features:

- Agent Engine **Code Execution** , now in Preview, lets your agent run code in an isolated sandbox environment. For more information, see [Code Execution](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/code-execution/overview).

- You can now develop, deploy, and use agents that support the **Agent-to-Agent (A2A) protocol** on Agent Engine. For more information, see [Develop an Agent2Agent agent](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/develop/a2a).

- Agent Engine now supports **bidirectional streaming** . For more information, see [Bidirectional streaming](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/bidirectional-streaming).

- The [Agent Engine page](https://console.cloud.google.com/vertex-ai/agents/agent-engines) in the Cloud Console UI now has a new **Memory Bank** tab for displaying and managing memories.

**Vertex AI** v1 Breaking **Vertex AI Agent Engine**

In version `v1.112.0` of the Vertex AI SDK for Python, the `agent_engines` module has been refactored to a [client-based design](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/overview#migration). For information about updating your existing code to the new design, see the [Migration guide](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/deprecations/agent-engine-migration).
Feature **Vertex AI Agent Engine**

Agent Engine now supports the following features:

- Agent Engine **Code Execution** , now in Preview, lets your agent run code in an isolated sandbox environment. For more information, see [Code Execution](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/code-execution/overview).

- You can now develop, deploy, and use agents that support the **Agent-to-Agent (A2A) protocol** on Agent Engine. For more information, see [Develop an Agent2Agent agent](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/develop/a2a).

- Agent Engine now supports **bidirectional streaming** . For more information, see [Bidirectional streaming](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/bidirectional-streaming).

- The [Vertex AI Agent Engine page](https://console.cloud.google.com/vertex-ai/agents/agent-engines) in the Google Cloud console has a **Memory Bank** tab for displaying and managing memories.

## September 09, 2025

**Generative AI on Vertex AI** v1 Feature [EmbeddingGemma](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/embeddinggemma) and [DeepSeek-V3.1](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-v3-1) models are available through Model Garden.
Feature [AI Singapore's SEA-LION V4 models](https://console.cloud.google.com/vertex-ai/publishers/aisingapore/model-garden/gemma-sea-lion-v4) are available through Model Garden. They are open models for Southeast Asian languages, built by leveraging Vertex Model Development Service for enhanced training efficiency and model accuracy.
**Vertex AI** v1 Feature [EmbeddingGemma](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/embeddinggemma) and [DeepSeek-V3.1](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-v3-1) models are available through Model Garden.
Feature [AI Singapore's SEA-LION V4 models](https://console.cloud.google.com/vertex-ai/publishers/aisingapore/model-garden/gemma-sea-lion-v4) are available through Model Garden. They are open models for Southeast Asian languages, built by leveraging Vertex AI training clusters for enhanced training efficiency and model accuracy.

## September 08, 2025

**Generative AI on Vertex AI** v1 Feature **Veo video generation**

Veo 3 support for short-duration videos is [generally available](https://cloud.google.com/products#product-launch-stages). You can use Veo 3 to create 4, 6, or 8 second videos. For more information, see the following:

- [Generate videos with Veo on Vertex AI from text prompts](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-text)
- [Generate videos with Veo on Vertex AI from an image](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-an-image)
- [Veo on Vertex AI video generation API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/veo-video-generation)

## September 03, 2025

**Generative AI on Vertex AI** v1 Change **Vertex AI RAG Engine: Managed Database (Spanner)**

Customers will be charged for the use of a Google-managed Spanner instance that's provisioned in a Google tenant project, using standard Spanner SKUs.

For more information, see [Vertex AI RAG Engine billing](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/rag-engine/rag-engine-billing).

## August 29, 2025

**Agent Platform Workbench** v2 Feature **M132 release**

The M132 release of Vertex AI Workbench instances includes the following:

- The new scheduler Jupyter plugin (`scheduler-jupyter-plugin`) is now preinstalled in the Jupyterlab 4 environment, with support for both the Cloud Composer and Vertex AI notebook schedulers.

- Updated the Dataproc JupyterLab plugin (`dataproc-jupyter-plugin`) to version 0.1.90.

- Patched bugs related to the managed end user credentials feature (Preview), resolving an incompatibility with listing Dataproc remote kernels.

- Patched a bug that caused instances with disabled proxy access to get stuck in provisioning.

- Removed the archived Debian 11 backports repository, resolving an issue with running `apt update` within the instance.

Change

### M132 Release

Change The new scheduler Jupyter plugin (`scheduler-jupyter-plugin`) is now
preinstalled in the Jupyterlab 4 environment, with support for both the Cloud
Composer and Vertex AI notebook schedulers.
Change Updated the Dataproc JupyterLab plugin (`dataproc-jupyter-plugin`) to version
0.1.90.
Fixed Patched bugs related to the managed end user credentials feature (Preview),
resolving an incompatibility with listing Dataproc remote kernels.
Fixed Patched a bug that caused instances with disabled proxy access to get stuck in
provisioning.
Change Removed the archived Debian 11 backports repository, resolving an issue with
running `apt update` within the instance.

## August 26, 2025

**Generative AI on Vertex AI** v1 Announcement **Gemini 2.5 Flash Image Preview**

[Gemini 2.5 Flash Image](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash#image-generation) (`gemini-2.5-flash-image-preview`) is available in [Preview](https://cloud.google.com/products#product-launch-stages). Gemini 2.5 Flash Image Preview supports additional image generation and editing features such as [image generation from multiple reference images](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/image-generation#image-generation) and [improved multi-turn image editing](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/image-editing#multi-turn-editing).
Feature **Vertex AI model tuning and Gen AI evaluation service**

Vertex AI model tuning now supports integration with the Gen AI evaluation service in Preview. You can automatically run evaluations on your tuned models and intermediate checkpoints. For more information, see [Create a tuning job](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini-use-supervised-tuning#create_a_text_model_supervised_tuning_job).

## August 21, 2025

**Generative AI on Vertex AI** v1 Feature **Vertex AI Agent Engine**

Agent Engine now supports the following enterprise security features:

- You can now deploy your agents in a private VPC environment, configuring a Private Service Connect interface, to ensure data privacy and meet security and compliance requirements. For more information, see [Configure Private Service Connect interface](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/deploy#psc-i).

- You can now use your own [customer-managed encryption keys (CMEK)](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/manage/access#cmek) to protect data at rest.

- You can now specify [customized resource controls](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/deploy#customized-resource-controls), such as the minimum and maximum number of application instances, resource limits for each container, and concurrency for each container.

- As a part of Vertex AI Platform, Vertex AI Agent Engine now supports [HIPAA](https://cloud.google.com/security/compliance/hipaa) workloads.

For more information, see [Agent Engine overview](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/overview#enterprise_security).
**Vertex AI** v1 Feature **Vertex AI Agent Engine**

Agent Engine now supports the following enterprise security features:

- You can now deploy your agents in a private VPC environment, configuring a Private Service Connect interface, to ensure data privacy and meet security and compliance requirements. For more information, see [Configure Private Service Connect interface](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/deploy#psc-i).

- You can now use your own [customer-managed encryption keys (CMEK)](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/manage/access#cmek) to protect data at rest.

- You can now specify [customized resource controls](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/deploy#customized-resource-controls), such as the minimum and maximum number of application instances, resource limits for each container, and concurrency for each container.

- As a part of Vertex AI, Vertex AI Agent Engine supports [HIPAA](https://cloud.google.com/security/compliance/hipaa) workloads.

For more information, see [Agent Engine overview](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/overview#enterprise_security).

## August 14, 2025

**Generative AI on Vertex AI** v1 Announcement **Imagen**

Imagen 4 is Generally Available.

Imagen 4 introduces the following models:

- [Imagen 4 Generate](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/imagen/4-0-generate-001)
- [Imagen 4 Fast Generate](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/imagen/4-0-fast-generate-001)
- [Imagen 4 Ultra Generate](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/imagen/4-0-ultra-generate-001)

For more information, see [Generate images using text prompts](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/generate-images.md.txt) and [Image generation API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/imagen-api).
Feature [Gemma 3 270M](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma3;publisherModelVersion=gemma-3-270m-it), [Wan 2.2](https://console.cloud.google.com/vertex-ai/publishers/wan-ai/model-garden/wan2-2) and [Wan 2.1](https://console.cloud.google.com/vertex-ai/publishers/wan-ai/model-garden/wan2-1) models are available through Model Garden.
**Vertex AI** v1 Feature [Gemma 3 270M](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma3;publisherModelVersion=gemma-3-270m-it), [Wan 2.2](https://console.cloud.google.com/vertex-ai/publishers/wan-ai/model-garden/wan2-2) and [Wan 2.1](https://console.cloud.google.com/vertex-ai/publishers/wan-ai/model-garden/wan2-1) models are available through Model Garden.

## August 13, 2025

**Generative AI on Vertex AI** v1 Feature [Qwen3 Coder](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/qwen/qwen3-coder) and [Qwen3 235B](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/qwen/qwen3-235b) are available as Model as a Service (MaaS) models in Model Garden.
Feature OpenAI's [gpt-oss-120b](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/openai/gpt-oss-120b) and [gpt-oss-20b](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/openai/gpt-oss-20b) are available as Model as a Service (MaaS) models in Model Garden.
**Vertex AI** v1 Feature [Qwen3 Coder](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/qwen/qwen3-coder) and [Qwen3 235B](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/qwen/qwen3-235b) are available as Model as a Service (MaaS) models in Model Garden.
Feature OpenAI's [gpt-oss-120b](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/openai/gpt-oss-120b) and [gpt-oss-20b](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/maas/openai/gpt-oss-20b) are available as Model as a Service (MaaS) models in Model Garden.

## August 08, 2025

**Generative AI on Vertex AI** v1 Feature **Gemini 2.5 Flash-Lite** and **Gemini 2.5 Pro** now support supervised fine-tuning. For more information, see [About supervised fine-tuning for Gemini models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini-supervised-tuning).

## August 07, 2025

**Generative AI on Vertex AI** v1 Feature **Vertex AI Agent Engine**

You can use your own [custom service account](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/set-up#custom-service-account) for agent identity to manage permissions and access according to your organization's security policies.
Feature **Model tuning**

You can now perform supervised fine-tuning on open models such as Llama 3.1. For more information, see [Tune an open model](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/open-model-tuning).
Feature **Vertex AI prompt optimizer**

The Vertex AI prompt optimizer is now [generally available](https://cloud.google.com/products?e=13802955&hl=en#product-launch-stages). For more information, see [Optimize prompts](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/prompts/prompt-optimizer).

We now offer a [zero-shot prompt optimizer](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/prompts/prompt-optimizer).

## August 06, 2025

**Generative AI on Vertex AI** v1 Feature [OpenAI's gpt-oss](https://console.cloud.google.com/vertex-ai/publishers/openai/model-garden/gpt-oss) models are available through Model Garden.
Feature **Imagen**

Virtual try-on lets you generate virtual try-on images from an image of a
person and product photos that you provide, and is available in Preview. For more information, see [Generate Virtual Try-On Images](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/generate-virtual-try-on-images.md.txt) and [Virtual Try-On API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/virtual-try-on-api).

This release note is incorrect; see entry for [October 9, 2025](https://docs.cloud.google.com/vertex-ai/docs/release-notes.md.txt#October_09_2025).
**Vertex AI** v1 Feature [OpenAI's gpt-oss](https://console.cloud.google.com/vertex-ai/publishers/openai/model-garden/gpt-oss) models are available through Model Garden.

## August 05, 2025

**Agent Platform Workbench** v2 Feature [Generally available](https://cloud.google.com/products#product-launch-stages): You can consume reservations with Vertex AI Workbench instances. Reservations of Compute Engine zonal resources help you gain a high level of assurance that your jobs have the necessary resources to run. For more information, see [Use reservations with Vertex AI Workbench instances](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/reservations).
**Colab Enterprise** Feature [Generally available](https://cloud.google.com/products#product-launch-stages): You can consume reservations with Colab Enterprise runtimes. Reservations of Compute Engine zonal resources help you gain a high level of assurance that your runtimes have the necessary resources to run. For more information, see [Use reservations with Colab Enterprise](https://docs.cloud.google.com/colab/docs/reservations.md.txt).

## August 04, 2025

**Colab Enterprise** Feature You can now use the new Data Science Agent to automate exploratory data analysis, perform machine learning tasks, and deliver insights from within a Colab Enterprise notebook. To get started, see [Use the Data Science Agent](https://docs.cloud.google.com/colab/docs/use-data-science-agent.md.txt). This feature is in [Preview](https://cloud.google.com/products#product-launch-stages).

## July 29, 2025

**Generative AI on Vertex AI** v1 Announcement **Veo video generation** Veo 3 and Veo 3 Fast are now generally available. For more information, see [Generate videos using text prompts](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-text).

## July 23, 2025

**Generative AI on Vertex AI** v1 Change [Grounding with Google Maps](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/grounding/grounding-with-google-maps) is available in all regions (except for the EEA) as a Preview (Pre-GA) feature.

## July 22, 2025

**Generative AI on Vertex AI** v1 Announcement **Gemini 2.5 Flash-Lite** is now generally available and accessible using the API and Vertex AI Studio. This GA release includes support for explicit caching and [batch prediction](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/batch-prediction-gemini), as well as expanded region support.

See [Gemini 2.5 Flash-Lite](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash-lite) for more information.

## July 17, 2025

**Generative AI on Vertex AI** v1 Announcement **Veo 3** preview models now support upscaling for 1080p resolution using the new `resolution` parameter. For more information, see [Veo on Vertex AI](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/veo-video-generation#resolution).

## July 16, 2025

**Generative AI on Vertex AI** v1 Feature Added [Gemma 3](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_axolotl_gemma3_finetuning.ipynb) fine-tuning notebook using Axolotl docker with support for 1b, 4b, 12b, and 27b variants.
**Vertex AI** v1 Feature Added [Gemma 3](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_axolotl_gemma3_finetuning.ipynb) fine-tuning notebook using Axolotl docker with support for 1b, 4b, 12b, and 27b variants.

## July 14, 2025

**Generative AI on Vertex AI** v1 Feature [Multimodal MedGemma 27B IT](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/medgemma;publisherModelVersion=medgemma-27b-it), [MedSigLIP](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/medsiglip), and [T5Gemma](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/t5gemma) models are available through Model Garden.
**Vertex AI** v1 Feature [Multimodal MedGemma 27B IT](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/medgemma;publisherModelVersion=medgemma-27b-it), [MedSigLIP](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/medsiglip), and [T5Gemma](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/t5gemma) models are available through Model Garden.

## July 11, 2025

**Vertex AI** v1 Feature To reduce the cost of running your inference jobs, you can use flex-start VMs, which are powered by [Dynamic Workload Scheduler](https://cloud.google.com/blog/products/compute/introducing-dynamic-workload-scheduler). Flex-start VMs offer significant discounts and are well-suited for
short-duration workloads. This feature is available in [Preview](https://cloud.google.com/products/#product-launch-stages).

For more information, see [Use DWS flex-start VMs with inference](https://docs.cloud.google.com/vertex-ai/docs/predictions/use-flex-start-vms).

## July 10, 2025

**Agent Platform Workbench** v2 Feature **M131 release**

The M131 release of Vertex AI Workbench instances includes the following:

- Updated the Dataproc JupyterLab plugin to version 0.1.89.
Change

### M131 Release

Change Updated the Dataproc JupyterLab plugin to version 0.1.89.

## July 08, 2025

**Generative AI on Vertex AI** v1 Feature **Vertex AI Agent Engine**

[Vertex AI Agent Engine Memory Bank](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/memory-bank/overview) is now available in Preview. Memory Bank lets you dynamically generate long-term memories based on users' conversations with your agent.

## July 03, 2025

**Generative AI on Vertex AI** v1 Feature **Vertex AI Agent Garden**

[Vertex AI Agent Garden](https://console.cloud.google.com/vertex-ai/agents/agent-garden) now supports filtering by tags.

## July 01, 2025

**Vertex AI** v1 Feature The global endpoint is generally available (GA) for Anthropic's Claude Opus 4. For details, see [Global endpoint](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/locations#global-endpoint).

## June 30, 2025

**Colab Enterprise** Feature [Preview](https://cloud.google.com/products#product-launch-stages): You can consume reservations with Colab Enterprise runtimes. Reservations of Compute Engine zonal resources help you gain a high level of assurance that your runtimes have the necessary resources to run. For more information, see [Use reservations with Colab Enterprise](https://docs.cloud.google.com/colab/docs/reservations.md.txt).
**Vertex AI** v1 Deprecated Mistral Nemo, which is offered as a Model as a Service (MaaS) model in Model Garden, is deprecated. For details, see [Model as a Service (MaaS) deprecations](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/deprecations/partner-models).
Deprecated Anthropic's Claude 3 Opus, which is offered as a Model as a Service (MaaS) model in Model Garden, is deprecated. For details, see [Model as a Service (MaaS) deprecations](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/deprecations/partner-models).
Feature Vertex AI now offers GA support of Private Service Connect Interface and includes Private DNS Peering. For more information, see [Use Private Service Connect interface for Vertex AI Training](https://docs.cloud.google.com/vertex-ai/docs/training/psc-i-egress).
Feature Private Service Connect interface (PSC-I) support for ML pipeline runs in Vertex AI Pipelines is now [generally available](https://cloud.google.com/products#product-launch-stages). PSC-I is recommended for private connectivity because it reduces the chance of IP exhaustion, allows for transitive peering, and includes Private DNS Peering.

For more information, see [Configure Private Service Connect interface for a pipeline](https://docs.cloud.google.com/vertex-ai/docs/pipelines/configure-private-service-connect).
Feature Vertex AI online inference now offers Preview support of PSC service automation that can automatically create PSC endpoints for dedicated private endpoints. For more information, see [Create the online inference endpoint with PSC automation](https://docs.cloud.google.com/vertex-ai/docs/predictions/private-service-connect#create-endpoint-with-automation).

## June 27, 2025

**Generative AI on Vertex AI** v1 Feature [Gemma 3n](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma3n) models are now available through Model Garden.
Feature Multimodal datasets are now available in preview. For more information, see [Multimodal datasets](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/datasets).
**Vertex AI** v1 Feature [Gemma 3n](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma3n) models are available through Model Garden.

## June 26, 2025

**Agent Platform Workbench** v2 Feature **M130 release**

The M130 release of Vertex AI Workbench instances includes the following:

- Updated the Dataproc JupyterLab plugin to version 0.1.87.
- Added the BigQuery JupyterLab plugin, version 0.0.1.
- The `GOOGLE_CLOUD_REGION` environment variable is now set by default.
Change

### M130 Release

Change Updated the Dataproc JupyterLab plugin to version 0.1.87.
Feature Added the BigQuery JupyterLab plugin, version 0.0.1.
Change The `GOOGLE_CLOUD_REGION` environment variable is now set by default.

## June 24, 2025

**Generative AI on Vertex AI** v1 Deprecated Starting on June 24, 2025, Imagen versions 1 and 2, image captioning, and visual question answering are deprecated.

On September 24, 2025, the following features and models will be removed:

- image captioning
- visual question answering
- Imagen 1 model `imagegeneration@002`
- Imagen 2 models `imagegeneration@005` and `imagegeneration@006`

For more information, see [Migrate to Imagen 3](https://docs.cloud.google.com/release-notes/vertex-ai/generative-ai/docs/image/migrate-to-imagen-3).
**Vertex AI** v1 Deprecated The Generative AI module in the Vertex AI SDK is deprecated. For information about migrating to the Google Gen AI SDK, see the [migration guide](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/deprecations/genai-vertexai-sdk.md.txt).

## June 23, 2025

**Generative AI on Vertex AI** v1 Announcement **Veo 2** support for advanced video controls is Generally Available. In addition to a providing a first frame of a video, you can specify the last frame of a video or a video to extend in length. For more information, see [Veo on Vertex AI API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/veo-video-generation).

## June 17, 2025

**Generative AI on Vertex AI** v1 Change **Updated pricing for Gemini 2.5 Flash GA:** The price for Gemini 2.5 Flash in GA will be adjusted to reflect its quality and unified output token pricing. This includes lower prices for thinking output, higher prices for non-thinking output. These pricing changes will take effect on the new GA endpoint as shared above. Preview pricing will only continue on existing preview endpoints for 30 days post-GA on **July 15, 2025**.
Change **Provisioned Throughput (PT):** Once a model is GA, all new PT purchases will be for GA endpoints only. If you've purchased PT for a specific preview version, it will still work for that specific preview. However, you must [migrate the existing PT to the GA endpoint](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/provisioned-throughput/purchase-provisioned-throughput#change-order) or purchase new PT for the GA endpoint by **July 15, 2025**.
Announcement **Gemini 2.5 Flash-Lite** is now available as a preview offering in both the API and Vertex AI Studio.

See [Gemini 2.5 Flash-Lite](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash-lite) for more information.
Announcement **Live API** is now available as a private general availability offering in the API and Vertex AI Studio. Reach out to your Google account team representative to request access.

See [Live API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/live-api) for more information.
Announcement **Gemini 2.5 Flash** and **Gemini 2.5 Pro** are now generally available and accessible using the API and Vertex AI Studio.

See [Gemini 2.5 Flash](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash) and [Gemini 2.5 Pro](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-pro) for more information.
Deprecated **Preview endpoint availability and removal:** All existing Gemini 2.5 Flash and Pro preview endpoints (listed below) will continue to be available with their current preview pricing until July 15, 2025. After this date, these preview endpoints will be shut down.

- `gemini-2.5-flash-preview-04-17`
- `gemini-2.5-flash-preview-05-20`
- `gemini-2.5-pro-preview-03-25`
- `gemini-2.5-pro-preview-05-06`
- `gemini-2.5-pro-preview-06-05`
Change **Updated preview endpoints:** Effective **June 19, 2025** , `gemini-2.5-flash-preview-04-17` endpoint will serve the Gemini 2.5 Flash model version released on 05-20, which has been promoted to GA. Similarly, the `gemini-2.5-pro-preview-05-06` and `03-25` endpoints will serve the Gemini 2.5 Pro model version released on 06-05, also promoted to GA. This update ensures continuity during your transition.

## June 16, 2025

**Generative AI on Vertex AI** v1 Announcement The DeepSeek API service on Vertex AI is in [Preview](https://cloud.google.com/products#product-launch-stages). For more information, see the [DeepSeek model card](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-r1-0528-maas) in Model Garden.

## June 11, 2025

**Generative AI on Vertex AI** v1 Change **Imagen 4**'s public preview models are updated to the following:

- `imagen-4.0-generate-preview-06-06`
- `imagen-4.0-fast-generate-preview-06-06`
- `imagen-4.0-ultra-generate-preview-06-06`

For more information about each model, see [Preview Imagen models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models#preview-imagen-models).

To avoid service interruption, migrate from `imagen-4.0-ultra-generate-exp-05-20` and `imagen-4.0-generate-preview-05-20` before 2025-07-07.

## June 10, 2025

**Agent Platform Workbench** v2 Feature Available in [Preview](https://cloud.google.com/products#product-launch-stages): You can consume reservations with Vertex AI Workbench instances. Reservations of Compute Engine zonal resources help you gain a high level of assurance that your jobs have the necessary resources to run. For more information, see [Use reservations with Vertex AI Workbench instances](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/reservations).
**Vertex AI** v1 Feature **Vector Search custom constraints with Organization Policy**

You can use custom constraints with the Organization Policy Service to provide more granular control over specific fields for indexes and index endpoints in Vector Search.

For more information, see [Create custom constraints for Vector Search](https://docs.cloud.google.com/vertex-ai/docs/vector-search/custom-constraints).

## June 09, 2025

**Generative AI on Vertex AI** v1 Change **Gemini API**

The `logprobs` and `response_logprobs` parameters for the Gemini API are now [generally available](https://cloud.google.com/products#product-launch-stages). For more information, see [Generate content with Gemini API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/inference#logprobs).

## June 05, 2025

**Generative AI on Vertex AI** v1 Feature **Gemini 2.5 Pro** 's public preview version has been updated to `gemini-2.5-pro-preview-06-05` and includes expanded support for thinking. This model version is available in the API and Vertex AI Studio.

See [Gemini 2.5 Pro](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-pro) for model details.

## June 03, 2025

**Generative AI on Vertex AI** v1 Announcement In Model Garden, the following fine tuning features have been added:

- [Gemma 3](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma3) UI fine-tuning using PEFT docker.
- [Qwen 2.5](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_qwen2_5_finetuning.ipynb) fine-tuning notebook using PEFT docker.
- [Qwen 3](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_axolotl_qwen3_finetuning.ipynb) fine-tuning notebook using Axolotl docker.
- lm-evaluation-harness as an evaluation service in the [Llama 3.3](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_llama3_3_finetuning.ipynb), [Llama 3.1](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_llama3_1_finetuning.ipynb), [Gemma 3](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_gemma3_finetuning_on_vertex.ipynb) and [Gemma 2](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_gemma2_finetuning_on_vertex.ipynb) fine-tuning notebooks.
Announcement Model Garden now includes [DeepSeek-R1-0528](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-r1) variants.

## May 28, 2025

**Colab Enterprise** Feature Python 3.11 is now available in Colab Enterprise. Existing runtimes and runtime templates will remain using Python 3.10. For more information, see [Python versions](https://docs.cloud.google.com/colab/docs/runtimes.md.txt#python).
Feature When you create a runtime template, you can now configure it to use the latest Python version available to Colab Enterprise, or you can specify the Python version. Using `Latest` is a new option that means when a new version of Python is introduced to Colab Enterprise, runtimes that you create will use the latest Python version.

Existing runtime templates and runtimes remain using their current Python version (Python 3.10). This includes existing auto-generated default runtime templates. To create default runtime templates that use `Latest`, you must do one of the following:

- Delete the existing default runtime templates. Then, when a new default runtime template is created, the Python version will be set to Latest.
- [Change a runtime template's Python version](https://docs.cloud.google.com/colab/docs/runtimes.md.txt#change-python-version) by using the REST API.

## May 23, 2025

**Generative AI on Vertex AI** v1 Announcement Mistral OCR is an Optical Character Recognition API for document understanding. It is [GA](https://cloud.google.com/products#product-launch-stages) on Vertex AI. For more information, see the [Mistral OCR model card](https://console.cloud.google.com/vertex-ai/publishers/mistralai/model-garden/mistral-ocr-2505) in Model Garden.

## May 22, 2025

**Generative AI on Vertex AI** v1 Announcement Anthropic's Claude Opus 4 and Claude Sonnet 4 are [GA](https://cloud.google.com/products#product-launch-stages) on Vertex AI and support [Provision Throughput](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/supported-models#partner-models). For more information, see the [Claude Opus 4](https://console.cloud.google.com/vertex-ai/publishers/anthropic/model-garden/claude-opus-4) or [Claude Sonnet 4](https://console.cloud.google.com/vertex-ai/publishers/anthropic/model-garden/claude-sonnet-4) model card in Model Garden.

## May 20, 2025

**Generative AI on Vertex AI** v1 Feature **Vertex AI Agent Engine**

The following features are now available in Preview:

- [Console support for managing deployed agents](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/manage/overview)
- [Sessions support for Agent Development Kit agents](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/sessions/overview)
Feature **Audio-to-audio** support for [**Gemini 2.5 Flash with Live API**](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash#live-api-a2a) is now available as a private preview. Users must be allowlisted to use this new feature.

The model is available in the API and Vertex AI Studio.

See [Live API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/live-api#a2a) for details.
Change **Gemini 2.5 Flash** 's public preview version has been updated to `gemini-2.5-flash-preview-5-20`.

See [Gemini 2.5 Flash](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash) for model details.

The model is available in the API and Vertex AI Studio.
Announcement **New stable text embeddings models** are now generally available:

- `gemini-embedding-001`
- `text-embedding-005`

For more information, see [Get text embeddings](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/embeddings/get-text-embeddings).
Feature **Thought summaries** are now available as an experimental feature for Gemini 2.5 Pro and 2.5 Flash.

For details, see [Thinking](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/thinking#thought-summaries).

The model is available in the API and Vertex AI Studio.
Feature **Imagen 4**

Imagen 4 offers two Preview models: [Imagen 4 Generate Preview 05-20](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/imagen/4-0-generate-preview-05-20), and [Imagen 4 Ultra Generate Experimental 05-20](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/imagen/4-0-ultra-generate-exp-05-20).

For more information, see
[Generate images using text prompts](http://cloud.google.com/vertex-ai/generative-ai/docs/image/generate-images) and the
[Generate images](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/imagen-api) API.

The model is available in the API and Vertex AI Studio.
Announcement **Lyria 2**, our latest music generation model, is now generally available.

See our [music generation prompt guide](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/music/music-gen-prompt-guide) and our [user guide](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/music/generate-music) for more information.

The model is available in the API and Vertex AI Studio.
Announcement [MedGemma](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/medgemma) models are available in Model Garden.
Feature **Veo 3**

[Veo 3](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/veo/3-0-generate-preview) is available in Preview for allowlisted accounts.

For more information about Veo 3, see [Veo \| AI Video Generator](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos) and [Veo on Vertex AI API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/veo-video-generation).

The model is available in the API and Vertex AI Studio.

## May 15, 2025

**Vertex AI** v1 Announcement Vertex AI custom training supports the following accelerators:

- A3 Ultra
- A4

Vertex AI online prediction supports the following accelerators:

- A3 Mega
- A4

For regional availability of accelerators, see [Using accelerators](https://docs.cloud.google.com/vertex-ai/docs/general/locations#accelerators).

## May 14, 2025

**Generative AI on Vertex AI** v1 Deprecated [MedLM](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/medlm/overview.md.txt) is deprecated. Access to MedLM will no longer be available on or after September 29, 2025.

## May 07, 2025

**Generative AI on Vertex AI** v1 Feature **Gemini 2.0 Flash with image generation** (`gemini-2.0-flash-preview-image-generation`) is now available as a public preview offering.

For more information, see [Generate images with Gemini](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/image-generation).
Fixed Seed parameter is now in GA and supports Gemini 2.5 model family.

## May 05, 2025

**Generative AI on Vertex AI** v1 Change **Grounding**

The following grounding features are [generally available](https://cloud.google.com/products#product-launch-stages):

- [Grounding on your data using Vertex AI Search](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/grounding/grounding-with-your-data)
- [Grounding with Elasticsearch](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/grounding/grounding-with-elasticsearch)

## May 02, 2025

**Generative AI on Vertex AI** v1 Announcement The global endpoint is generally available ([GA](https://cloud.google.com/products#product-launch-stages)). For details, see [Global endpoint](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/locations#global-endpoint).

## April 30, 2025

**Generative AI on Vertex AI** v1 Change
- Additional materials are available for deploying a model in Model Garden by using the [Python SDK, gcloud CLI, or API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-garden/use-models#deploy_a_model), which are available in [Preview](https://cloud.google.com/products#product-launch-stages):
  - [Developer blog: Introducing the new Vertex AI Model Garden CLI and SDK](https://www.googlecloudcommunity.com/gc/Community-Blogs/Introducing-the-new-Vertex-AI-Model-Garden-CLI-and-SDK/ba-p/888386)
  - [Deploy open models by using the SDK tutorial notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_deployment_tutorial.ipynb)
  - [Get started with Vertex AI Model Garden SDK notebook](https://github.com/GoogleCloudPlatform/generative-ai/blob/main/open-models/get_started_with_model_garden_sdk.ipynb)
Feature
- Llama 4 Maverick and Scout models are available in [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden) with [Model-as-a-Service API Service](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/llama-4-maverick-17b-128e-instruct-maas) and [self-hosted deployments](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/llama4).
- [HiDream-I1](https://console.cloud.google.com/vertex-ai/publishers/hidream-ai/model-garden/hidream-i1), [Llama Guard 4](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/llama-guard), [Llama Prompt Guard 2](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/prompt-guard), and [Qwen3](https://console.cloud.google.com/vertex-ai/publishers/qwen/model-garden/qwen3) are available in [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden).

## April 29, 2025

**Colab Enterprise** Feature Gemini in Colab Enterprise, which is a product in the Gemini for Google Cloud portfolio, now includes additional capabilities in [Preview](https://cloud.google.com/products/#product-launch-stages). See the following:

- [Chat about your notebook with Gemini](https://docs.cloud.google.com/colab/docs/chat.md.txt)
- [Explain code cells with Gemini assistance](https://docs.cloud.google.com/colab/docs/explain-cells.md.txt)
- [Explain and fix errors with Gemini assistance](https://docs.cloud.google.com/colab/docs/explain-errors.md.txt)

To enable and activate Gemini in Colab Enterprise features, see [Set up Gemini in Colab Enterprise](https://docs.cloud.google.com/gemini/docs/colab/set-up-gemini).
**Generative AI on Vertex AI** v1 Announcement Gemini 1.5 Pro and Gemini 1.5 Flash models are not available in projects that have no prior usage of these models, including new projects. For details, see
[Model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/model-versions).

## April 21, 2025

**Colab Enterprise** Feature **The notebook gallery is now available.**

The notebook gallery is a curated collection of notebooks to help you get started using Colab Enterprise. This collection consists of ready-to-use templates and examples to make it easier to learn new techniques, understand best practices, and get projects started quickly. Browse the notebooks by category or use the search bar to find a notebook that helps you get started. See the [notebook gallery](https://console.cloud.google.com/vertex-ai/colab/notebook-gallery).

## April 17, 2025

**Generative AI on Vertex AI** v1 Announcement [Gemini 2.5 Flash](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-flash) with thinking and other well-rounded capabilities is now available in [Preview](https://cloud.google.com/products#product-launch-stages).

## April 16, 2025

**Agent Platform Workbench** v2 Feature **M129 release**

The M129 release of Vertex AI Workbench instances includes the following:

- Updated the Dataproc JupyterLab plugin to version 0.1.85.
Change

### M129 Release

Change Updated the Dataproc JupyterLab plugin to version 0.1.85.
**Vertex AI** v1 Announcement [Persistent resources](https://docs.cloud.google.com/vertex-ai/docs/training/persistent-resource-overview) for custom training is generally available ([GA](https://cloud.google.com/products#product-launch-stages)) and supports rebooting.

## April 10, 2025

**Generative AI on Vertex AI** v1 Announcement Managed APIs for Llama 4 Maverick and Scout are in [Preview](https://cloud.google.com/products#product-launch-stages) on Vertex AI. For more information, see the [Llama 4 model card](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/llama-4-maverick-17b-128e-instruct-maas).

## April 09, 2025

**Generative AI on Vertex AI** v1 Feature **Vertex AI Agent Engine**

The following features are now available for Vertex AI Agent Engine in Preview:

- [Agent Development Kit integration](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/develop/adk)
- [Example Store](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/example-store/overview)
- [LlamaIndex Query Pipeline integration](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/develop/llama-index/query-pipeline)

The following features are now generally available for Vertex AI Agent Engine:

- [Agent monitoring](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/manage/monitoring)
Feature **Gemini Live API** is now available as a public preview offering and has been updated with the following features:

- Support for responses in 8 voices and 31 languages using Chirp 3
- Updated UI support in Vertex AI Studio
- Expanded conversation session window
- Ability to extend conversation sessions
- Support to share your current screen with Gemini during conversations
- Transcription support for audio in and audio out
- Support to change or update the system instructions mid-session

For more information, see [Gemini 2.0 Flash Live API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/live-api).
Feature **Agent Development Kit** (ADK) is now available in [Preview](https://cloud.google.com/products#product-launch-stages). For more information, see [Agent Development Kit](https://google.github.io/adk-docs/).
Feature **Gemini 2.5 Pro** is now available as a public preview offering.

For more information, see [Gemini 2.5 Pro](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-pro).
Feature **Agent Garden** is now available in [Preview](https://cloud.google.com/products#product-launch-stages). For more information, see [Vertex AI Agent Builder overview](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-builder/overview) or go directly to [Agent Garden](https://console.cloud.google.com/vertex-ai/agents/agent-garden) in the Cloud Console.
Change **Vertex AI Agent Builder** now refers to a suite of features for building and deploying AI agents in Vertex AI. For more information see, [Vertex AI Agent Builder overview](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-builder/overview).

The original Vertex AI Agent Builder product has been renamed **AI Applications** . The product functionality and endpoints remain the same. For more information, see [What is AI Applications?](https://docs.cloud.google.com/generative-ai-app-builder/docs/introduction).
Feature **Grounding: Web Grounding for Enterprise** is now Generally available. For more information, see [Web Grounding for Enterprise](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/grounding/web-grounding-enterprise).
Feature **Grounding: Grounding with Google Maps** is now available as a Public Experimental feature. For more information, see [Grounding with Google Maps](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/grounding/grounding-with-google-maps).

## March 31, 2025

**Colab Enterprise** Feature [Preview](https://cloud.google.com/products#product-launch-stages): You can switch to a default runtime with GPUs by using a button in your Colab Enterprise notebook. To enable a default runtime with GPUs for your users, see [Enable default runtimes with GPUs](https://docs.cloud.google.com/colab/docs/default-runtimes-with-gpus.md.txt).

## March 27, 2025

**Vertex AI** v1 Feature **Generally available**: To reduce the cost of running your training and prediction jobs, you can use Spot VMs. Spot VMs are virtual machine (VM) instances that are excess Compute Engine capacity. Spot VMs have significant discounts, but Compute Engine might preemptively stop or delete Spot VMs to reclaim the capacity at any time.

For more information, see [Use Spot VMs with training](https://docs.cloud.google.com/vertex-ai/docs/training/use-spot-vms) and [Use Spot VMs with prediction](https://docs.cloud.google.com/vertex-ai/docs/predictions/use-spot-vms).

## March 26, 2025

**Agent Platform Workbench** v2 Feature The ability to back up and restore data on a Vertex AI Workbench instance is now [generally available](https://cloud.google.com/products#product-launch-stages). For more information, see [Back up and restore data on an instance](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/restore).
**Vertex AI** v1 Feature **Generally available**: You can consume reservations of VMs that have GPUs attached with your custom training jobs or prediction jobs. Reservations of Compute Engine zonal resources help you gain a high level of assurance that your jobs have the necessary resources to run. For more information, see the following:

- [Use reservations with training](https://docs.cloud.google.com/vertex-ai/docs/training/use-reservations)

- [Use reservations with prediction](https://docs.cloud.google.com/vertex-ai/docs/predictions/use-reservations)

## March 25, 2025

**Generative AI on Vertex AI** v1 Feature
- [DeepSeek-V3-0324](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-v3), [TxGemma](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/txgemma) and [Sesame CSM](https://console.cloud.google.com/vertex-ai/publishers/sesame/model-garden/csm) are now available in [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden).
- [DeepSeek-R1](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-r1), [V3](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-v3) and [V3-0324](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-v3) can be deployed with H200 GPUs and improved vLLM support.
- You can deploy a model in Model Garden by using the [Python SDK, gcloud CLI, or API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-garden/use-models#deploy_a_model), which are available in [Preview](https://cloud.google.com/products#product-launch-stages). You can get started with the "Equivalent code" in the deploy panel in the Model Garden console.

## March 20, 2025

**Agent Platform Workbench** v2 Feature Encrypt your data-in-use by using [Confidential Computing](https://docs.cloud.google.com/confidential-computing/docs/confidential-computing-overview.md.txt). This feature is now available in [Preview](https://cloud.google.com/products#product-launch-stages). You can enable the Confidential VM service when you create a Vertex AI Workbench instance. To get started, see [Create an instance with Confidential Computing](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/create-confidential-computing).
**Generative AI on Vertex AI** v1 Announcement Anthropic's Claude Sonnet 3.7 is [GA](https://cloud.google.com/products#product-launch-stages) on Vertex AI and supports [Provision Throughput](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/supported-models#partner-models). To learn more, view the [Claude Sonnet 3.7 model card](https://console.cloud.google.com/vertex-ai/publishers/anthropic/model-garden/claude-3-7-sonnet) in Model Garden.

## March 17, 2025

**Generative AI on Vertex AI** v1 Announcement Mistral Small 3.1 (25.03) feature multimodal capabilities and a context of up to 128,000 tokens. For more information, see the [Mistral Small 3.1 (25.03) model card](https://console.cloud.google.com/vertex-ai/publishers/mistralai/model-garden/mistral-small-2503) in Model Garden.

## March 14, 2025

**Generative AI on Vertex AI** v1 Feature [Judge model evaluation and customization tools](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/evaluate-judge-model) are now available in Preview for the [Gen AI evaluation service in Vertex AI](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/evaluation-overview).

## March 13, 2025

**Generative AI on Vertex AI** v1 Announcement [Context caching](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/context-cache/context-cache-overview) for Gemini on Vertex AI is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

## March 12, 2025

**Agent Platform Workbench** v2 Feature **M128 release**

The M128 release of Vertex AI Workbench instances includes the following:

- Miscellaneous package updates.
Change

### M128 Release

Change Miscellaneous package updates.
v1 Feature **M128 release**

The M128 release of Vertex AI Workbench user-managed notebooks includes the following:

- Miscellaneous package updates.
Feature The M128 release of Vertex AI Workbench managed notebooks includes the following:

- Miscellaneous package updates.
**Generative AI on Vertex AI** v1 Change Model Garden fine tuning updates:

- Added a [workbench-based notebook for Llama 3.1 finetuning](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_llama3_1_finetuning_with_workbench.ipynb).
- Updated [Llama 3.1](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_llama3_1_finetuning.ipynb) and [Gemma 2](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_gemma2_finetuning_on_vertex.ipynb) UI fine-tuning with the updated PEFT docker.
Feature
- [Gemma 3](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma3) and [ShieldGemma 2](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/shieldgemma2) are now available in Model Garden.
- [CogVideoX-2b](https://console.cloud.google.com/vertex-ai/publishers/thudm/model-garden/cogvideox) is now available in Model Garden.

## March 11, 2025

**Generative AI on Vertex AI** v1 Change **Gemini 2.0 Flash Tuning**

Gemini 2.0 Flash fine-tuning is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

Added support for [tuning function calling](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/tune-function-calling).

## March 04, 2025

**Generative AI on Vertex AI** v1 Announcement **Vertex AI Agent Engine**

[Vertex AI Agent Engine](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/overview) is now [generally available (GA)](https://cloud.google.com/products#product-launch-stages).

Billing for Vertex AI Agent Engine starts on March 4, 2025. We recommend that you delete unused resources to avoid incurring unwanted costs. For more information, see [Pricing](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/overview#pricing).
Change LangChain on Vertex AI has been renamed to [Vertex AI Agent Engine](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/overview).

## February 27, 2025

**Colab Enterprise** Feature You can use Terraform resources to schedule notebook runs, and to manage runtimes and runtime templates. To learn more, see the following:

- [Schedule a notebook run](https://docs.cloud.google.com/colab/docs/schedule-notebook-run.md.txt#run_notebook_once-terraform)
- [Create a runtime](https://docs.cloud.google.com/colab/docs/create-runtime.md.txt#colab-create-runtime-terraform)
- [Create a runtime template](https://docs.cloud.google.com/colab/docs/create-runtime-template.md.txt#colab-create-runtime-template-terraform)

## February 25, 2025

**Generative AI on Vertex AI** v1 Feature **Gemini 2.0 Flash-Lite is now generally available**

Gemini 2.0 Flash-Lite is now generally available. For more information, see [Gemini 2.0](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/gemini-v2).

## February 24, 2025

**Generative AI on Vertex AI** v1 Announcement Anthropic's Claude Sonnet 3.7 is in [Preview](https://cloud.google.com/products#product-launch-stages) on Vertex AI. To learn more, view the [Claude Sonnet 3.7 model card](https://console.cloud.google.com/vertex-ai/publishers/anthropic/model-garden/claude-3-7-sonnet) in Model Garden.

## February 21, 2025

**Generative AI on Vertex AI** v1 Change
- PEFT Docker updates
  - Added support for evaluation metrics like perplexity, bleu, google_bleu, rouge1, rouge2, rougeL, rougeLSum.
  - Uses the best checkpoint and loads the model based on the best eval metrics.
  - Run training and eval only for data which is less than or equal to the `max_seq_length`.
  - Use `gcloud storage rsync` instead of `csfuse` to save a checkpoint.
- Fine tuning updates
  - You can select a service account when you click **Fine-tune** for a model, such as [Llama 3.1](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/llama3_1).
  - Added a [PEFT based LLM finetuning tutorial notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_finetuning_tutorial.ipynb).
  - Added a [Axolotl based LLM finetuning notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_axolotl_finetuning.ipynb).
  - Updated [Llama 3.1](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_llama3_1_finetuning.ipynb) and [Gemma 2](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_gemma2_finetuning_on_vertex.ipynb) fine-tuning notebooks with the updated PEFT Docker container.
- Model updates
  - Updated the [PaliGemma](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/paligemma) model card by supporting PaliGemma 2 mix models, and segmentation functionality to Paligemma 1 models.
  - Updated the [LLaVa model card](https://console.cloud.google.com/vertex-ai/publishers/liuhaotian/model-garden/pytorch-llava) by supporting LLaVA Next models and adding vLLM to the notebook.

## February 12, 2025

**Generative AI on Vertex AI** v1 Feature Deepseek-V3 and Deepseek-R1 have been added to [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden) in [Preview](https://cloud.google.com/products#product-launch-stages):

- [DeepSeek-V3](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-v3) (671B) is a powerful Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.
- [DeepSeek-R1](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-r1) (671B) is one of the first-generation reasoning models introduced by DeepSeek and offers performance comparable to OpenAI-o1 across math, code, and reasoning tasks.

You can use a [notebook](https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_deepseek_deployment.ipynb) to deploy these models.

## February 11, 2025

**Generative AI on Vertex AI** v1 Announcement The [Llama 3.3 70B](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/llama-3.3-70b-instruct-maas) model that is managed on Vertex AI is now in [Preview](https://cloud.google.com/products#product-launch-stages).

## February 07, 2025

**Generative AI on Vertex AI** v1 Feature [deepseek-ai/deepseek-r1](https://console.cloud.google.com/vertex-ai/publishers/deepseek-ai/model-garden/deepseek-r1) and [microsoft/Phi-4](https://console.cloud.google.com/vertex-ai/publishers/microsoft/model-garden/phi4) models were added to [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden).
Change The following advanced LLM inference optimization techniques are available in Model Garden in [Preview](https://cloud.google.com/products#product-launch-stages):

- Prefix caching reuses computations from previously generated text, eliminating redundant processing. It reduces time-to-first-token for requests with common prompt prefixes. Prefix caching is available for the following models:

  - vLLM: Llama 3.1 (8b, 70b), Llama 3.3 (70b)
  - Hex-LLM: Llama 2 (7b, 13b), Llama 3 (8b), Llama 3.1 (8b, 70b), Llama 3.2 (1b, 3b), Llama Guard (1b, 8b), CodeLlama (7b, 13b), Gemma (2b, 7b), CodeGemma (2b, 7b), Mistral-7B (v0.2, v0.3), Mixtral-8x7B (v0.1)
- Speculative decoding is an effective optimization technique to reduce generation time-per-output-token latency. For more information, see the [Model Garden advanced features notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_advanced_features.ipynb).

## February 05, 2025

**Generative AI on Vertex AI** v1 Feature **Gemini 2.0 Flash general availability for text-only output**

Gemini 2.0 Flash is now generally available for text-only outputs. Multimodal outputs are still available only as a private preview. For more information, see [Gemini 2.0](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/gemini-v2).
Feature **New Gemini 2.0 Pro and Gemini 2.0 Flash-Lite models available to users**

Two new models in the Gemini 2.0 family are now available to users:

- **Gemini 2.0 Pro:** Our strongest model for coding and world knowledge, featuring a 2M long context window. Gemini 2.0 Pro is available as an experimental model in Vertex AI.
- **Gemini 2.0 Flash-Lite:** Our fastest and most cost efficient Flash model. Gemini 2.0 Flash-Lite is available as a Preview model in Vertex AI.

For more information, see [Gemini 2.0](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/gemini-v2)

## January 31, 2025

**Generative AI on Vertex AI** v1 Feature You can now monitor usage, throughput, and latency and troubleshoot 429 errors on Vertex AI foundation models, like Google Gemini and Anthropic Claude, by using a predefined dashboard. After querying a model from the [Vertex AI Model Garden](https://cloud.google.com/model-garden), you can find the name of the model you queried in the Vertex AI **Dashboard** page under the "Model observability" heading.

To customize the dashboard and explore relevant metrics in Cloud Monitoring, click **Show All Metrics** . For information about using dashboards in Cloud Monitoring, see [View and customize Google Cloud dashboards](https://docs.cloud.google.com/monitoring/charts/predefined-dashboards.md.txt).
**Vertex AI** v1 Feature [Dedicated Public Endpoints](https://docs.cloud.google.com/vertex-ai/docs/general/deployment#deploy_a_model_to_an_endpoint)and [Private Service Connect Endpoints (PSC-E)](https://docs.cloud.google.com/vertex-ai/docs/predictions/private-service-connect) for Vertex AI Prediction are generally available (GA).

## January 30, 2025

**Generative AI on Vertex AI** v1 Deprecated Mistral Large (24.07) and Codestral (24.05) that are offered as a Model as a Service (MaaS) models in Model Garden are deprecated. For details, see [Generative AI on Vertex AI deprecations](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/deprecations.md.txt).

## January 29, 2025

**Generative AI on Vertex AI** v1 Feature **New Imagen 3 image generation model available to users**

A newer improved Imagen 3 image generation model is now available to all users:

- `imagen-3.0-generate-002`

This image generation model supports the following additional features:

- **Prompt enhancement** - The LLM-based prompt rewriter tool adds additional details and descriptive language to the prompt you provide, generally resulting in higher quality generated images. This feature is configurable and is enabled by default.

For more information, see [Imagen on Vertex AI model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/model-versioning) and [Generate images using text prompts](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/generate-images.md.txt).

## January 22, 2025

**Generative AI on Vertex AI** v1 Announcement **LangChain on Vertex AI**

Billing for LangChain on Vertex AI will start on March 4, 2025.

The pricing structure is based on vCPU hours and GiB hours used. This means that you will be charged for both the compute (vCPU) and memory resources consumed by your LangChain on Vertex AI workloads.

You can review the pricing details in the table below.

| Product | SKU ID | Price |
|---|---|---|
| ReasoningEngine vCPU | 8A55-0B95-B7DC | $0.0994/vCPU-Hr |
| ReasoningEngine Memory | 0B45-6103-6EC1 | $0.0105/GiB-Hr |

## January 21, 2025

**Generative AI on Vertex AI** v1 Deprecated Anthropic's Claude 3 Sonnet that is offered as a Model as a Service (MaaS) model in Model Garden is deprecated. For details, see [Generative AI on Vertex AI deprecations](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/deprecations.md.txt).

## January 17, 2025

**Generative AI on Vertex AI** v1 Feature [Agent evaluation](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/evaluation-agents) using the [Gen AI evaluation service](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/evaluation-overview) is available in [Preview](https://cloud.google.com/products#product-launch-stages).

## January 16, 2025

**Agent Platform Workbench** v2 Feature **M127 release**

The M127 release of Vertex AI Workbench instances includes the following:

- Fixed an issue related to ownership of the home directory when using authorized ssh keys.
Change

### M127 Release

Fixed Fixed an issue related to ownership of the home directory when using authorized
ssh keys.
v1 Feature **M127 release**

The M127 release of Vertex AI Workbench user-managed notebooks includes the following:

- Fixed an issue related to ownership of the home directory when using authorized ssh keys.
Feature The M127 release of Vertex AI Workbench managed notebooks includes the following:

- Fixed an issue related to ownership of the home directory when using authorized ssh keys.

## December 20, 2024

**Generative AI on Vertex AI** v1 Change [RAG Engine](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/rag-overview) is [generally available (GA)](https://cloud.google.com/products#product-launch-stages).

The supported models include the following:

- Google Gemini
- Google embedding and OSS E5 embedding models
- Model Garden self-deployed OSS LLMs
- [Model as a service (MaaS) Llama models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/use-partner-models)

The supported features include the following:

- **[Data connectors](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/use-data-connectors)**: Google Cloud Storage, Google Drive, Slack, Jira, and SharePoint
- **[Document types](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/supported-documents)**: Google Workspace documents, HTML, JSON, Markdown, PDF, and text files
- **Transformations**: fixed-size chunking and chunk overlap
- **Vector databases**: Vertex AI Vector Search and Pinecone
**Vertex AI** v1 Feature **Vector Search hybrid search and sparse embeddings are generally available (GA)**

Vector Search hybrid search and sparse embeddings are generally available ([GA](https://cloud.google.com/products/#product-launch-stages)).
Hybrid search uses both dense and sparse embeddings, which lets you search based on a combination of keyword search and semantic search. For more information about hybrid search, see [About hybrid search](https://docs.cloud.google.com/vertex-ai/docs/vector-search/about-hybrid-search).

## December 18, 2024

**Generative AI on Vertex AI** v1 Feature [Hex-LLM: High-Efficiency Large Language Model Serving](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/open-models/use-hex-llm) is available in [General Availability (GA)](https://cloud.google.com/products#product-launch-stages).

This launch adds support for the following models:

- Llama 3.1
- Llama 3.2
- Phi-3
- Qwen2 and Qwen2.5

Additional supported features:

- Multi-host serving.
- Disaggregated serving (experimental).
- Prefix caching.
- AWQ quantization.

## December 17, 2024

**Generative AI on Vertex AI** v1 Feature You can copy tuned Gemini 1.5 Pro 002 and Gemini 1.5 Flash 002 adapter models across projects. For details, see [Copy a model in Vertex AI Model Registry](https://docs.cloud.google.com/vertex-ai/docs/model-registry/copy-model).
**Vertex AI** v1 Feature You can copy tuned Gemini 1.5 Pro 002 and Gemini 1.5 Flash 002 adapter models across projects. For details, see [Copy a model in Vertex AI Model Registry](https://docs.cloud.google.com/vertex-ai/docs/model-registry/copy-model).

## December 11, 2024

**Generative AI on Vertex AI** v1 Feature The Gemini 2.0 Flash (`gemini-2.0-flash-exp`) model is Generally available for grounded answer generation with RAG. This model is tuned to address context-based question and answering tasks. For more information, see [Ground responses for Gemini models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/ground-gemini).

## December 10, 2024

**Colab Enterprise** Feature Gemini in Colab Enterprise, which is a product in the Gemini for Google Cloud portfolio, now includes error fixing in [Preview](https://cloud.google.com/products/#product-launch-stages). Gemini in Colab Enterprise can suggest fixes when your code produces errors. For more information, see [Fix errors](https://docs.cloud.google.com/colab/docs/use-code-completion.md.txt#fix-errors).

To enable and activate Gemini in Colab Enterprise features, see [Set up Gemini in Colab Enterprise](https://docs.cloud.google.com/gemini/docs/colab/set-up-gemini).
**Generative AI on Vertex AI** v1 Feature **Imagen 3 image generation models Generally Available to all users**

Imagen 3 image generation models are now available to all users without requiring prior approval. These include the following image generation models:

- `imagen-3.0-generate-001`
- `imagen-3.0-fast-generate-001` (low latency model)

Prior image generation models (`imagegeneration@006`, `imagegeneration@005`, `imagegeneration@002`) still require approval to use.

For more information, see [Imagen on Vertex AI model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/model-versioning) and [Generate images using text prompts](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/generate-images.md.txt).
Feature **Imagen 3 Customization model Generally Available to approved users**

Imagen 3 Customization model is now available to approved users. This includes the following model:

- `imagen-3.0-capability`

Imagen 3 Customization lets you guide image generation by providing reference images (few-shot learning). Imagen 3 Customization lets you customize generated images for the following feature categories:

- [Subject Customization](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/subject-customization.md.txt) (product, person, and animal companion)
- [Style Customization](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/style-customization.md.txt)
- [Controlled Customization](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/edit-controlled.md.txt) (canny edge and scribble)
- [Instruct Customization](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/instruct-customization.md.txt) (Style transfer)
Feature **Imagen 3 editing model Generally Available to approved users**

The Imagen 3 Editing model is now available to approved users. This includes the following model:

- `imagen-3.0-capability`

This model offers the following additional features:

- [Inpainting](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/edit-inpainting) - Add or remove content from a masked area of an image
- [Outpainting](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/edit-outpainting.md.txt) - Expand a masked area of an image
- [Product image editing](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/edit-product-image) - Identify and maintain a primary product while changing the background or product position

For more information, see [Model versions](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/model-versioning).

## December 06, 2024

**Generative AI on Vertex AI** v1 Security A vulnerability was discovered in the Vertex AI API serving Gemini multimodal requests, allowing bypass of VPC Service Controls. For details, see the [Security bulletins](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/security-bulletins#gcp-2024-063) page.

## December 02, 2024

**Vertex AI** v1 Feature **[Preview](https://cloud.google.com/products/#product-launch-stages)**: You can consume reservations of VMs that have GPUs attached with your custom training jobs or prediction jobs. Reservations of Compute Engine zonal resources help you gain a high level of assurance that your jobs have the necessary resources to run. For more information, see the following:

- [Use reservations with training](https://docs.cloud.google.com/vertex-ai/docs/training/use-reservations)

- [Use reservations with prediction](https://docs.cloud.google.com/vertex-ai/docs/predictions/use-reservations)

## November 21, 2024

**Generative AI on Vertex AI** v1 Feature The Gen AI evaluation service can now help you evaluate your translation models using MetricX, COMET, and BLEU metrics.
To learn more about evaluating your translation models, see [Evaluate translation models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/determine-eval#translation-models).
Announcement Mistral Large (24.11) is Generally Available on Vertex AI as a managed model. To learn more, view the [Mistral Large (24.11) model card](https://console.cloud.google.com/vertex-ai/publishers/mistralai/model-garden/mistral-large-2411) in Model Garden.

## November 20, 2024

**Agent Platform Workbench** v2 Feature **M126 release**

The M126 release of Vertex AI Workbench instances includes the following:

- [Preview](https://cloud.google.com/products#product-launch-stages): JupyterLab 4+ is available on new Vertex AI Workbench instances. To try it, select JupyterLab 4 when you [create your instance](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/create#create).
- Upgraded JupyterLab to 3.6.8.
Change

### M126 Release

Feature [Preview](https://cloud.google.com/products#product-launch-stages): JupyterLab
4+ is available on new Vertex AI Workbench instances. To try it, select
JupyterLab 4 when you [create your
instance](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/create#create).
Change Upgraded JupyterLab to 3.6.8.
v1 Feature The M126 release of Vertex AI Workbench managed notebooks includes the following:

- Upgraded JupyterLab to 3.6.8.
Feature **M126 release**

The M126 release of Vertex AI Workbench user-managed notebooks includes the following:

- Upgraded JupyterLab to 3.6.8.
- One or more framework versions have reached their end of patch and support dates. To view end of patch and support dates, see [Supported framework versions](https://docs.cloud.google.com/vertex-ai/docs/workbench/user-managed/images#supported-frameworks).

## November 08, 2024

**Generative AI on Vertex AI** v1 Feature [Batch predictions for Llama models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/llama-batch) on Vertex AI (MaaS) is available in [Preview](https://cloud.google.com/products#product-launch-stages).
Feature **Batch prediction support for Gemini**

Batch prediction is available for Gemini in [General Availability (GA)](https://cloud.google.com/products#product-launch-stages). Available Gemini models include Gemini 1.0 Pro, Gemini 1.5 Pro, and Gemini 1.5 Flash. To get started with batch prediction, see [Get batch predictions for Gemini](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/batch-prediction-gemini).

## November 05, 2024

**Generative AI on Vertex AI** v1 Change We are extending the availability of Gemini 1.0 Pro 001 and Gemini 1.0 Pro Vision 001 from February 15, 2025 to April 9, 2025. For details, see the [Deprecations](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/deprecations.md.txt).

## November 04, 2024

**Generative AI on Vertex AI** v1 Announcement The Anthropic Claude Haiku 3.5 is Generally Available on Vertex AI. To learn more, view the [Claude Haiku 3.5 model card](https://console.cloud.google.com/vertex-ai/publishers/anthropic/model-garden/claude-3-5-haiku) in Model Garden.
Change The translation LLM now supports Polish, Turkish, Indonesian, Dutch, Vietnamese, Thai and Czech. For the full list of supported languages, see the [Translate text](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/translate/translate-text) page.

## October 31, 2024

**Vertex AI** v1 Feature PSC-I Egress is supported for Ray clusters Vertex AI. PSC-I is recommended for private connectivity since it reduces the chance of IP exhaustion, and allows for transitive peering. Check out [Private Service Connect interface for Ray on Vertex AI](https://docs.cloud.google.com/vertex-ai/docs/open-source/ray-on-vertex-ai/create-cluster#use-psc-i-egress). This feature is available in [Preview](https://cloud.google.com/products#product-launch-stages).
Feature Private Service Connect interface (PSC-I) is now supported for ML pipeline runs in Vertex AI Pipelines. PSC-I is recommended for private connectivity, since it reduces the chance of IP exhaustion, and allows for transitive peering.

For more information, see [Configure Private Service Connect interface for a pipeline](https://docs.cloud.google.com/vertex-ai/docs/pipelines/configure-private-service-connect). This feature is available in [Preview](https://cloud.google.com/products#product-launch-stages).

## October 28, 2024

**Generative AI on Vertex AI** v1 Feature The [Whisper large v3 and Whisper large v3 turbo](https://console.cloud.google.com/vertex-ai/publishers/openai/model-garden/whisper-large) models have been added to Model Garden.
Feature You can now fine-tune the following models from the Cloud console:

- [Gemma 2](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma2)
- [Llama 3.1](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/llama3_1)
- [Mistral](https://console.cloud.google.com/vertex-ai/publishers/mistral-ai/model-garden/mistral)
- [Mixtral](https://console.cloud.google.com/vertex-ai/publishers/mistral-ai/model-garden/mixtral)
Change Updated the fine-tuning notebooks for [Gemma 2](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma2), [Llama 3.1](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/llama3_1), [Mistral](https://console.cloud.google.com/vertex-ai/publishers/mistral-ai/model-garden/mistral), and [Mixtral](https://console.cloud.google.com/vertex-ai/publishers/mistral-ai/model-garden/mixtral) with the following enhancements:

- The notebooks use an updated high-performance container for single host multi-GPU LoRA fine-tuning.
  - Better throughput and GPU utilization with well-tested max-sequence-lengths.
  - Support for input token masking.
  - No out of memory (OOM) error during fine-tuning.
- Added a custom dataset example that uses a template and format validation.
- Support for a default accelerator pool with quota checks.
- Improved documentation.

## October 25, 2024

**Colab Enterprise** Feature Colab Enterprise is now available in the following regions:

- Hamina, Finland (`europe-north1`)
- Milan, Italy (`europe-west8`)
- Tel Aviv, Israel (`me-west1`)
- Warsaw, Poland (`europe-central2`)

See [Colab Enterprise locations](https://docs.cloud.google.com/colab/docs/locations.md.txt).

## October 22, 2024

**Generative AI on Vertex AI** v1 Announcement The Anthropic Claude Sonnet 3.5 v2 is Generally Available. To learn more, view the [Claude Sonnet 3.5 v2 model card](https://console.cloud.google.com/vertex-ai/publishers/anthropic/model-garden/claude-3-5-sonnet-v2) in Model Garden.

## October 18, 2024

**Generative AI on Vertex AI** v1 Announcement The [Llama 3.1 405B](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/llama-3.1-405b-instruct-maas) model that is managed on Vertex AI is now [Generally Available](https://cloud.google.com/products#product-launch-stages).

## October 09, 2024

**Generative AI on Vertex AI** v1 Feature The Vertex AI Gemini API SDK supports tokenization capabilities for local token counting and computation. This is a streamlined way to compute tokens locally, ensuring compatibility across different Gemini models and their tokenizers. Supported models include ` gemini-1.5-flash` and ` gemini-1.5-pro` . To learn more, see [Count tokens](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/list-token).

## October 08, 2024

**Vertex AI** v1 Feature **Vector Search Private Service Connect automation**

Deploying an index with Private Service Connect automation is generally available ([GA](https://cloud.google.com/products#product-launch-stages)). You can set up a service connection policy so that you don't have to manually create a compute address and forwarding rule after each index deployment.

For more information, see [Set up Vector Search with Private Service Connect](https://docs.cloud.google.com/vertex-ai/docs/vector-search/setup/private-service-connect).

## October 04, 2024

**Generative AI on Vertex AI** v1 Change Added dynamic LoRA serving for [Llama 3.1](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_llama3_1_deployment.ipynb) and [Stable Diffusion XL](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_stable_diffusion_xl_lora.ipynb).
Feature You can deploy Hugging Face models on Google Cloud that have [text embedding inference](https://huggingface.github.io/text-embeddings-inference/) enabled or [pytorch inference](https://huggingface.co/docs/inference-endpoints/supported_tasks)
enabled. For more information, see the [Hugging Face model deployment](http://console.cloud.google.com/vertex-ai/model-garden;action=deploy;hfSource=true) in the console.
Feature [Prompt Guard](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/prompt-guard) and [Flux](https://console.cloud.google.com/vertex-ai/publishers/black-forest-labs/model-garden/flux1-schnell) were added to [Model Garden](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-garden/explore-models).
Change Added multiple deployment settings (with A100-80G and H100) and sample requests for some popular models, including [Llama 3.1](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/llama3_1), [Gemma 2](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma2), and [Mixtral](https://console.cloud.google.com/vertex-ai/publishers/mistral-ai/model-garden/mixtral).
Feature The AI assistant in Vertex AI Studio can help you refine and generate prompts. This feature is in [Preview](https://cloud.google.com/products#product-launch-stages). To learn more, see [Use AI-powered prompt writing tools](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/prompts/ai-powered-prompt-writing).

## October 03, 2024

**Colab Enterprise** Feature Gemini in Colab Enterprise, which is a product in the Gemini for Google Cloud portfolio, is [generally available](https://cloud.google.com/products/#product-launch-stages). Gemini in Colab Enterprise helps you write code by suggesting code as you type. You can also use the Help me code tool to generate code from a description of what you want.

Gemini in Colab Enterprise is available to try at no cost through December 31, 2024.

To learn how to enable and activate Gemini in Colab Enterprise features, see [Set up Gemini in Colab Enterprise](https://docs.cloud.google.com/gemini/docs/colab/set-up-gemini).

## October 01, 2024

**Generative AI on Vertex AI** v1 Feature **Grounding: Dynamic retrieval for grounded results (GA)**

Dynamic retrieval lets you choose when to turn off grounding with Google Search. This is useful when a prompt doesn't require an answer grounded in Google Search, and the supported models can provide an answer based on their knowledge without grounding. Dynamic retrieval helps you manage latency, quality, and cost more effectively.

This feature is [Generally Available](https://cloud.google.com/products#product-launch-stages). For more information, see [Dynamic retrieval](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/ground-gemini#dynamic-retrieval).

## September 30, 2024

**Generative AI on Vertex AI** v1 Feature Prompt templates let you to test how different prompt formats perform with different sets of prompt data. This feature is in [Preview](https://cloud.google.com/products#product-launch-stages). To learn more, see [Use prompt templates](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/prompts/prompt-templates).

## September 26, 2024

**Agent Platform Workbench** v2 Feature **M125 release**

The M125 release of Vertex AI Workbench instances includes the following:

- `bigframes` 1.9.0 is now available in all environments except TensorFlow.
- Fixed a regression introduced in M124 where Conda was getting downgraded to an older version.
- Patched a vulnerability with `adm` and `docker` permissions when the instance's root access isn't enabled.
Change

### M125 Release

Feature `bigframes` 1.9.0 is now available in all environments except TensorFlow.
Fixed Fixed a regression introduced in M124 where Conda was getting downgraded to an
older version.
Fixed Patched a vulnerability with `adm` and `docker` permissions when the instance's
root access isn't enabled.
v1 Feature **M125 release**

The M125 release of Vertex AI Workbench user-managed notebooks includes the following:

- Patched a vulnerability with `adm` and `docker` permissions when the instance's root access isn't enabled.
Feature The M125 release of Vertex AI Workbench managed notebooks includes the following:

- Patched a vulnerability with `adm` and `docker` permissions when the instance's root access isn't enabled.

## September 25, 2024

**Generative AI on Vertex AI** v1 Announcement The Llama 3.2 90B model is available in [Preview](https://cloud.google.com/products#product-launch-stages) on Vertex AI. Llama 3.2 90B enables developers to build and deploy the latest generative AI models and applications that use Llama's capabilities, such as image reasoning. Llama 3.2 is also designed to be more accessible for on-device applications. For more information, see [Llama models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/llama).

## September 24, 2024

**Generative AI on Vertex AI** v1 Announcement New stable versions of Gemini 1.5 Pro (`gemini-1.5-pro-002`) and Gemini 1.5 Flash (`gemini-1.5-flash-002`)
are [Generally Available](https://cloud.google.com/products#product-launch-stages). These models introduce broad quality improvements over the previous `001` versions, with significant gains in the following categories:

- Factuality and reduce model hallucinations
- Openbook Q\&A for RAG use cases
- Instruction following
- Multilingual understanding in 102 languages, especially in Korean, French, German, Spanish, Japanese, Russian, and Chinese.
- SQL generation
- Audio understanding
- Document understanding
- Long context
- Math and reasoning

For more information about differences with the previous model versions, see [Model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/model-versions).
Feature Use Gemini to directly analyze YouTube videos and publicly available media (such as images, audio, and video) by using a link. This feature is in [Public Preview](https://cloud.google.com/products#product-launch-stages).
Announcement The 2M context window with Gemini 1.5 Pro is now in [Generally Available](https://cloud.google.com/products#product-launch-stages), which opens up long-form multimodal use cases that only Gemini can support.
Feature The new API parameters `audioTimestamp`, `responseLogprob`, and `logprobs` are in [Public Preview](https://cloud.google.com/products#product-launch-stages). For more information, see [API reference](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/inference).
Feature The Vertex AI prompt optimizer adapts your prompts using the optimal instructions and examples to elicit the best performance from your chosen model. This feature is available in [Preview](https://cloud.google.com/products#product-launch-stages). To learn more, see [Optimize prompts](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/prompts/prompt-optimizer).
Change Gemini 1.5 Pro and Gemini 1.5 Flash Tuning is now available in [GA](https://cloud.google.com/products#product-launch-stages).
Tune Gemini with text, image, audio, and document data types using the latest models:

- `gemini-1.5-pro-002`
- `gemini-1.5-flash-002`

Gemini 1.0 tuning remains in preview.

For more information on tuning Gemini, see [Tune Gemini models by using supervised fine-tuning](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini-use-supervised-tuning).
Feature Gemini 1.5 Pro and Gemini 1.5 Flash now support multimodal input with [function calling](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/function-calling). This feature is in [Preview](https://cloud.google.com/products#product-launch-stages).
Change The latest versions of **Gemini 1.5 Flash** (`gemini-1.5-flash-002`) and **Gemini 1.5 Pro** (`gemini-1.5-pro-002`) use [dynamic shared quota](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/dsq), which distributes on-demand capacity among all queries being processed. Dynamic shared quota is [Generally Available](https://cloud.google.com/products#product-launch-stages).
Announcement [Controlled generation](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/control-generated-output) is now [Generally Available](https://cloud.google.com/products#product-launch-stages).

## September 23, 2024

**Colab Enterprise** Feature You can now use customer-managed encryption keys (CMEK) to protect notebooks in Colab Enterprise.

For more information, see [Use customer-managed encryption keys](https://docs.cloud.google.com/colab/docs/cmek.md.txt).

## September 20, 2024

**Generative AI on Vertex AI** v1 Feature Add label metadata to `generateContent` and `streamGenerateContent` API calls. For details, see [Add labels to API calls](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/add-labels-to-api-calls).

## September 18, 2024

**Generative AI on Vertex AI** v1 Announcement Model Garden supports an organization policy so that administrators can limit access to certain models and capabilities. For more information, see [Control access to Model Garden models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/control-model-access)

## September 17, 2024

**Vertex AI** v1 Feature To ensure that VM resources are available when your custom training and prediction jobs need them, you can now use Compute Engine reservations. Reservations provide a high level of assurance in obtaining capacity for Compute Engine resources. This feature is available in [Preview](https://cloud.google.com/products/#product-launch-stages) for A2 and A3 machine series reservations.

For more information, see [Use reservations with training](https://docs.cloud.google.com/vertex-ai/docs/training/use-reservations) and [Use reservations with prediction](https://docs.cloud.google.com/vertex-ai/docs/predictions/use-reservations).
Feature To reduce the cost of running your training and prediction jobs, you can now use Spot VMs. Spot VMs are virtual machine (VM) instances that are excess Compute Engine capacity. Spot VMs have significant discounts, but Compute Engine might preemptively stop or delete Spot VMs to reclaim the capacity at any time. This feature is available in [Preview](https://cloud.google.com/products/#product-launch-stages).

For more information, see [Use Spot VMs with training](https://docs.cloud.google.com/vertex-ai/docs/training/use-spot-vms) and [Use Spot VMs with prediction](https://docs.cloud.google.com/vertex-ai/docs/predictions/use-spot-vms).

## September 16, 2024

**Vertex AI** v1 Feature Schedule Vertex AI custom training jobs based on resource availability. For details, see the [Vertex AI documentation](https://docs.cloud.google.com/vertex-ai/docs/training/schedule-jobs-dws).

## September 10, 2024

**Agent Platform Workbench** v2 Feature The ability to back up and restore data on a Vertex AI Workbench instance is now available in [Preview](https://cloud.google.com/products#product-launch-stages). For more information, see [Back up and restore an instance](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/restore).

## September 09, 2024

**Vertex AI** v1 Feature Ray cluster's autoscaling feature is now supported. See [Scale Ray clusters on Vertex AI](https://docs.cloud.google.com/vertex-ai/docs/open-source/ray-on-vertex-ai/scale-clusters#autoscaling)

## September 03, 2024

**Generative AI on Vertex AI** v1 Change Gemini 1.5 Flash (`gemini-1.5-flash`) supports [controlled generation](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/control-generated-output).

## August 30, 2024

**Generative AI on Vertex AI** v1 Feature Gen AI Evaluation Service is Generally Available. To learn more, see the [Gen AI Evaluation Service overview](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/evaluation-overview).

## August 26, 2024

**Generative AI on Vertex AI** v1 Change For controlled generation, you can have the model respond with an enum value in plain text, as defined in your response schema. Set the `responseMimeType` to `text/x.enum`. For more information, see [Control generated output](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/control-generated-output).

## August 22, 2024

**Generative AI on Vertex AI** v1 Change **AI21 Labs**

Managed models from AI21 Labs are available on Vertex AI. To use a AI21 Labs model on Vertex AI, send a request directly to the Vertex AI API endpoint. For more information, see [AI21 models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/ai21).

## August 20, 2024

**Agent Platform Workbench** v1 Feature **M124 release**

The M124 release of Vertex AI Workbench user-managed notebooks includes the following:

- Pytorch 2.3.0 with CUDA 12.1 and Python 3.10 user-managed notebooks instances are now available.
- Fixed a bug that prevented kernels from appearing when the Cloud Resource Manager API is turned off and Dataproc is enabled.

## August 19, 2024

**Agent Platform Workbench** v2 Announcement The ability to create a Vertex AI Workbench instance based on a custom container is now [generally available](https://cloud.google.com/products#product-launch-stages). Only custom containers derived from the Google-provided base container are supported. For more information, see [Create an instance using a custom container](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/create-custom-container).

## August 14, 2024

**Colab Enterprise** Feature The notebook scheduler is now [generally available](https://cloud.google.com/products/#product-launch-stages). See [Schedule a notebook run](https://docs.cloud.google.com/colab/docs/schedule-notebook-run.md.txt).

## August 11, 2024

**Vertex AI** v1 Change Generative AI on Vertex AI supports CMEK, VPC Service Controls, and Data Residency. For more information, see [Security controls](https://docs.cloud.google.com/vertex-ai/docs/general/features#security_controls_support_on).

## August 09, 2024

**Generative AI on Vertex AI** v1 Announcement Gemini on Vertex AI supports multiple response candidates. For details, see [Generate content with the Gemini API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/inference).

## August 08, 2024

**Agent Platform Workbench** v2 Feature **M124 release**

The M124 release of Vertex AI Workbench instances includes the following:

- Fixed a bug that prevented kernels from appearing when the Cloud Resource Manager API is turned off and Dataproc is enabled.
- Spark notebooks on Dataproc: The Serverless Spark runtime template creation screen now has an easy-to-use UI for configuring resource allocation, autoscaling, and GPU settings.
Change

### M124 Release

Fixed Fixed a bug that prevented kernels from appearing when the Cloud Resource
Manager API is turned off and Dataproc is enabled.
Change Spark notebooks on Dataproc: The Serverless Spark runtime template creation
screen now has an easy-to-use UI for configuring resource allocation,
autoscaling, and GPU settings.
v1 Feature **M124 release**

The M124 release of Vertex AI Workbench managed notebooks includes the following:

- Fixed a bug that prevented kernels from appearing when the Cloud Resource Manager API is turned off and Dataproc is enabled.

## August 05, 2024

**Colab Enterprise** Fixed Fixed an issue in which users weren't able to access the Colab Enterprise UI when Colab Service Status was **OFF for everyone** in Google Workspace.
**Generative AI on Vertex AI** v1 Change The translation LLM now supports Arabic, Hindi, and Russian. For the full list of supported languages, see the [Translate text](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/translate/translate-text) page.

## August 02, 2024

**Generative AI on Vertex AI** v1 Feature Vertex AI SDK for Python supports token listing and counting for prompts without the need to make API calls. This feature is available in ([Preview](https://cloud.google.com/products#product-launch-stages)). For details, see [List and count tokens](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/list-token).
**Vertex AI** v1 Feature The Vertex AI Model Registry now offers [Preview](https://cloud.google.com/products/#product-launch-stages) support for model copy across different projects. For information about how to copy your model projects and regions, see [Copy models in Model Registry](https://docs.cloud.google.com/vertex-ai/docs/model-registry/copy-model).

## July 31, 2024

**Generative AI on Vertex AI** v1 Feature **New Imagen on Vertex AI image generation model and features**

The Imagen 3 image generation models (`imagen-3.0-generate-001` and the low-latency version `imagen-3.0-fast-generate-001`) are Generally Available to [approved users](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/overview.md.txt#feature-launch-stage). These models offer the following additional features:

- Additional aspect ratios (1:1, 3:4, 4:3, 9:16, 16:9)
- Digital watermark ([SynthID](https://deepmind.google/technologies/synthid/)) enabled by default
- Watermark verification
- User-configurable safety features (safety setting, person/face setting)

For more information, see [Model versions](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/model-versioning) and [Generate images using text prompts](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/generate-images.md.txt).
Announcement Gemma 2 2B is available in Model Garden. For details, see [Use Gemma open models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/open-models/use-gemma).
Feature The following models have been added to [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden):

- [Gemma 2 2B](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma2): A foundation LLM by Google DeepMind.
- [Qwen2](https://console.cloud.google.com/vertex-ai/publishers/qwen/model-garden/qwen2): An LLM series by Alibaba Cloud.
- [Phi-3](https://console.cloud.google.com/vertex-ai/publishers/microsoft/model-garden/phi3): An LLM series by Microsoft.
Change Resource and deployment settings were made to the following models:

- Added GPU inferences for [gemma2-27b](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma2) and [gemma2-27b-it](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma2) with verified performances.
- Added verified deployment settings for Mistral AI models that are deployed from Huggingface, including [mistralai/mistral-nemo-instruct-2407](https://console.cloud.google.com/vertex-ai/publishers/mistralai/model-garden/mistral-nemo-instruct-2407;action=deploy;hfSource=true), [mistralai/mistral-nemo-base-2407](https://console.cloud.google.com/vertex-ai/publishers/mistralai/model-garden/mistral-nemo-base-2407;action=deploy;hfSource=true), [mistralai/mistral-large-instruct-2407](https://console.cloud.google.com/vertex-ai/publishers/mistralai/model-garden/mistral-large-instruct-2407;action=deploy;hfSource=true), and [mistralai/codestral-22b-v0.1](https://console.cloud.google.com/vertex-ai/publishers/mistralai/model-garden/codestral-22b-v0.1;action=deploy;hfSource=true).
- Added multiple deployment settings with A100 (40G), A100 (80G) and H100 (80G) for select models, such as [llama3.1](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/llama3_1), [llama3](https://console.cloud.google.com/vertex-ai/publishers/meta/model-garden/llama3), [gemma2](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma2), [gemma](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma), and [mistral-7b](https://console.cloud.google.com/vertex-ai/publishers/mistral-ai/model-garden/mistral-7b).

## July 30, 2024

**Generative AI on Vertex AI** v1 Announcement See the [Gemini Online Inference on Vertex AI Service Level Agreement (SLA)](https://cloud.google.com/vertex-ai/generative-ai/sla).

## July 24, 2024

**Agent Platform Workbench** v1 Feature **M123 release**

The M123 release of Vertex AI Workbench managed notebooks includes the following:

- Fixed a bug that caused conflicting permissions with the Jupyter user and google-sudoers.
- Updated Nvidia drivers to version 550.90.07 to fix vulnerabilities.
**Generative AI on Vertex AI** v1 Announcement **Mistral AI**

Managed models from Mistral AI are available on Vertex AI. To use a Mistral AI model on Vertex AI, send a request directly to the Vertex AI API endpoint. For more information, see [Mistral AI models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/mistral).

## July 23, 2024

**Generative AI on Vertex AI** v1 Announcement **Llama 3.1**

The Llama 3.1 405B model is available in [Preview](https://cloud.google.com/products#product-launch-stages) on Vertex AI. Llama 3.1 405B provides capabilities from synthetic data generation to model distillation, steerability, math, tool use, multilingual translation, and more. For more information, see [Llama models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/llama).

## July 16, 2024

**Agent Platform Workbench** v2 Feature **M123 release**

The M123 release of Vertex AI Workbench instances includes the following:

- Fixed a bug that caused conflicting permissions with the Jupyter user and google-sudoers.
Change

### M123 Release

Fixed Fixed a bug that caused conflicting permissions with the Jupyter user and
google-sudoers.
v1 Feature **M123 release**

The M123 release of Vertex AI Workbench user-managed notebooks includes the following:

- Fixed a bug that caused conflicting permissions with the Jupyter user and google-sudoers.
- Fixed a bug for custom container instances using a disabled root.
**Colab Enterprise** Change All Colab Enterprise runtimes are automatically configured with a 100 GiB boot disk in addition to the disk specified in the runtime template. Starting July 16, 2024, the boot disk of a newly created Colab Enterprise runtime automatically defaults to an SSD Persistent Disk. Previously, the boot disk default was a Standard Persistent Disk.

Because of this change, default boot disks of Colab Enterprise runtimes are billed as SSD Persistent Disks instead of Standard Persistent Disks. For more information, see [Colab Enterprise pricing](https://cloud.google.com/colab/pricing).

## July 02, 2024

**Generative AI on Vertex AI** v1 Announcement Google's open weight Gemma 2 model is available in Model Garden. For details, see [Use Gemma open models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/open-models/use-gemma).
Change MaMMUT is now available in [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden). MaMMUT is a vision-encoder and text-decoder model for multimodal tasks such as visual question answering, image-text retrieval, text-image retrieval, and generation of multimodal embeddings.

## June 28, 2024

**Generative AI on Vertex AI** v1 Feature Launched [Hex-LLM](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/open-models/use-hex-llm) for high-efficiency large language model serving. This performant TPU serving solution is based on XLA and optimized kernels to achieve high throughput and low latency.

Hex-LLM uses several parallelism strategies for multiple TPU chips, quantizations, dynamic LoRA, and more. Hex-LLM supports the following dense and sparse LLMs:

- Gemma 2B and 7B
- Gemma 2 9B and 27B
- Llama 2 7B, 13B and 70B
- Llama 3 8B and 70B
- Mistral 7B and Mixtral 8x7B
Feature The following models have been added to [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden):

- 36 Hugging Face embedding models with verified deployment settings such as [BAAI/bge-m3](https://console.cloud.google.com/vertex-ai/publishers/BAAI/model-garden/bge-m3;action=deploy;hfSource=true) and [intfloat/multilingual-e5-large-instruct](https://console.cloud.google.com/vertex-ai/publishers/intfloat/model-garden/multilingual-e5-large-instruct;action=deploy;hfSource=true).
- 35 Hugging Face PyTorch models with verified deployment settings such as [stabilityai/stable-diffusion-2-1](https://console.cloud.google.com/vertex-ai/publishers/stabilityai/model-garden/stable-diffusion-2-1;action=deploy;hfSource=true).

For more information, see the [Hugging Face model deployment](http://console.cloud.google.com/vertex-ai/model-garden;action=deploy;hfSource=true) in the console.
Change
- Updated Docker images in [Llama 3 notebooks](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_llama3_finetuning.ipynb) that are more efficient at tuning.
- A notebook-based interactive workshop UI was added in [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden) for image generative models such as [stable-diffusion-xl-base](https://console.cloud.google.com/vertex-ai/publishers/stability-ai/model-garden/stable-diffusion-xl-base), [image inpainting](https://console.cloud.google.com/vertex-ai/publishers/runwayml/model-garden/stable-diffusion-inpainting), [controlnet](https://console.cloud.google.com/vertex-ai/publishers/lllyasviel/model-garden/control-net). You can find these models from the **Open Notebook** list.
- Colab Notebooks for frequently used models in Model Garden have been revised with no-code or low-code implementations to improve accessibility and user experience.
**Vertex AI** v1 Announcement Vertex AI custom training on TPU VMs support [customer managed encryption keys (CMEK)](https://docs.cloud.google.com/vertex-ai/docs/general/cmek).

## June 27, 2024

**Generative AI on Vertex AI** v1 Feature Context caching is available for Gemini 1.5 Pro. Use context caching to reduce the cost of requests that contain repeat content with high input token counts. For more information, see [Context caching overview](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/context-cache/context-cache-overview).

## June 25, 2024

**Generative AI on Vertex AI** v1 Feature Controlled generation is available on Gemini 1.5 Pro and supports the JSON schema. For more information, see [Control generated output](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/control-generated-output).

## June 21, 2024

**Agent Platform Workbench** v2 Feature **M122 release**

The M122 release of Vertex AI Workbench instances includes the following:

- Updated Nvidia drivers to version 550.90.07 to fix vulnerabilities.
Change

### M122 Release

Change Updated Nvidia drivers to version 550.90.07 to fix vulnerabilities.
v1 Feature **M122 release**

The M122 release of Vertex AI Workbench user-managed notebooks includes the following:

- Updated Nvidia drivers to version 550.90.07 to fix vulnerabilities.

## June 20, 2024

**Generative AI on Vertex AI** v1 Announcement The Anthropic Claude Sonnet 3.5 is Generally Available. To learn more, view the [Claude Sonnet 3.5 model card](https://console.cloud.google.com/vertex-ai/publishers/anthropic/model-garden/claude-3-5-sonnet) in Model Garden.
**Vertex AI** v1 Announcement Vertex AI custom training supports TPU v5e in `us-central1`. For details, see [Vertex AI locations](https://docs.cloud.google.com/vertex-ai/docs/general/locations).

## June 18, 2024

**Vertex AI** v1 Deprecated Starting on September 15, 2024, you can only customize
classification, entity extraction,
and sentiment analysis objectives by moving to Vertex AI Gemini prompts and tuning. Training or updating
models for Vertex AI AutoML for Text classification, entity extraction, and sentiment analysis
objectives will no longer be available. You can continue using existing Vertex AI AutoML Text models
until June 15, 2025. For more information about how Gemini offers enhanced user experience through
improved prompting capabilities, see [Overview of model tuning for
Gemini](https://cloud.google.com/vertex-ai/generative-ai/docs/models/tune-models).

## June 17, 2024

**Colab Enterprise** Feature You can now use customer-managed encryption keys (CMEK) to protect runtimes in Colab Enterprise. Using CMEK for notebook files isn't currently supported.

For more information, see [Use customer-managed encryption keys for runtimes](https://docs.cloud.google.com/colab/docs/cmek.md.txt).
**Generative AI on Vertex AI** v1 Change Increased the input token limit for Gemini 1.5 Pro from 1M to 2M. For more information, see [Google models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/models).

## June 11, 2024

**Generative AI on Vertex AI** v1 Change **Upload media from Google Drive**

You can upload media, such as PDF, MP4, WAV, and JPG files from Google Drive, when you send [image](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/image-understanding), [video](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/video-understanding), [audio](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/audio-understanding), and [document](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/document-understanding) prompt requests.

## June 10, 2024

**Colab Enterprise** Feature Gemini in Colab Enterprise, which is a product in the Gemini for Google Cloud portfolio, is available in [Preview](https://cloud.google.com/products/#product-launch-stages). Gemini in Colab Enterprise helps you write code by suggesting code as you type. You can also use the **Help me code** tool to generate code from a description of what you want.

To learn how to enable and activate Gemini in Colab Enterprise features, see [Set up Gemini in Colab Enterprise](https://docs.cloud.google.com/gemini/docs/colab/set-up-gemini).
Feature The notebook scheduler is now available in [Preview](https://cloud.google.com/products/#product-launch-stages). You can schedule a notebook to run immediately one time, or on a recurring schedule.

For more information, see [Schedule a notebook run](https://docs.cloud.google.com/colab/docs/schedule-notebook-run.md.txt).
**Generative AI on Vertex AI** v1 Feature **Experiment in the Vertex AI Studio login-free**

The Vertex AI Studio multi-model prompt designer can be accessed login-free. With this feature, prospective customers can use the Vertex AI Studio to test queries before deciding to sign up and create an account. To learn more about this experience, see [Vertex AI Studio console experiences](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/overview#console-experiences) or to access the console directly go to [Vertex AI Studio](https://console.cloud.google.com/vertex-ai/generative/multimodal/create/text?_ga=2.160908414.2102276437.1718032367-729345960.1717709985).

## June 07, 2024

**Agent Platform Workbench** v2 Feature You can now create a Vertex AI Workbench instance based on a custom container. This feature is available in [Preview](https://cloud.google.com/products#product-launch-stages). Only custom containers derived from the Google-provided base container are supported. For more information, see [Create an instance using a custom container](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/create-custom-container).

## June 03, 2024

**Agent Platform Workbench** v2 Feature You can now use [Workforce Identity Federation](https://docs.cloud.google.com/iam/docs/workforce-identity-federation.md.txt) with Vertex AI Workbench instances in [Preview](https://cloud.google.com/products/#product-launch-stages). Workforce Identity Federation lets you create and manage Vertex AI Workbench instances with credentials provided by an external identity provider (IdP).
For more information, see [Create an instance with third party credentials](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/create-third-party-instance).

## May 31, 2024

**Generative AI on Vertex AI** v1 Change **Generative AI on Vertex AI Regional APIs**

Generative AI on Vertex AI regional APIs are available in the [following three regions](https://cloud.google.com/vertex-ai/generative-ai/docs/learn/locations):

- `us-east5`
- `me-central1`
- `me-central2`

<br />

Change **Anthropic Claude 3.0 Opus model**

The [Anthropic Claude 3.0 Opus](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/use-claude#claude-opus) model is [Generally Available](https://cloud.google.com/products#product-launch-stages). To learn more, see its [model card in Model Garden](https://console.cloud.google.com/vertex-ai/publishers/anthropic/model-garden/claude-3-opus).
**Vertex AI** v1 Change **Vertex AI Regional APIs**

Vertex AI regional APIs are available in the [following seven regions](https://cloud.google.com/vertex-ai/docs/general/locations#available-regions):

- `us-east5`
- `us-south1`
- `africa-south1`
- `europe-southwest1`
- `europe-west12`
- `me-central1`
- `me-central2`

<br />

Feature Model Monitoring v2 is in [Preview](https://cloud.google.com/products#product-launch-stages), which centralizes model monitoring configuration and visualization on a model version and enables monitoring models being served outside of Vertex AI. For more information, see [Vertex AI Model Monitoring overview](https://docs.cloud.google.com/vertex-ai/docs/model-monitoring/overview).

## May 28, 2024

**Generative AI on Vertex AI** v1 Feature Gemini models support the `frequencyPenalty` and `presencePenalty` parameters. Use `frequencyPenalty` to control the probability of repeated text in a response. Use `presencePenalty` to control the probability of generating more diverse content. For more information, see [Gemini model parameters](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/gemini#parameters).
**Vertex AI** v1 Announcement **Vector Search sparse embeddings and hybrid search in Public preview**

Vector Search supports sparse embeddings and hybrid search in Public preview. Hybrid search uses both dense and sparse embeddings, which lets you search based on a combination of keyword search and semantic search. For how to format dense, sparse, and hybrid embeddings, see [Input data and structure](https://docs.cloud.google.com/vertex-ai/docs/vector-search/setup/format-structure).

## May 24, 2024

**Generative AI on Vertex AI** v1 Announcement The Gemini 1.5 Pro (`gemini-1.5-pro-001`) and Gemini 1.5 Flash (`gemini-1.5-flash-001`) models are [Generally Available](https://cloud.google.com/products?e=13802955&hl=en#product-launch-stages). For more information, see [Google models](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/models), [Overview of the Gemini API](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/overview), and [Send multimodal prompt requests](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/send-multimodal-prompts).

## May 20, 2024

**Generative AI on Vertex AI** v1 Feature The following models have been added to [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden):

- **E5**: A text embedding model series that can be served with a GPU or CPU.
- **Instant ID**: An identity preserving text-to-image generation model.
- **Stable Diffusion XL lightning**: A text-to-image generation model that is based on SDXL but requires fewer inference iterations.

To see a list of all available models, see [Explore models in Model Garden](https://docs.cloud.google.com/vertex-ai/docs/start/explore-models).

## May 17, 2024

**Agent Platform Workbench** v2 Feature **M121 release**

The M121 release of Vertex AI Workbench instances includes the following:

- Updated Nvidia drivers to 550.54.15 to fix an issue where Nvidia drivers failed to install on startup after Debian 11 images upgraded kernel to `linux-image-5.10.0-29-cloud-amd64`.
- The `linux-headers-cloud-amd64` metapackage is now installed for faster driver recompiling on kernel upgrades.
Change

### M121 Release

Change Updated Nvidia drivers to 550.54.15 to fix an issue where Nvidia drivers failed
to install on startup after Debian 11 images upgraded kernel to
`linux-image-5.10.0-29-cloud-amd64`.
Change The `linux-headers-cloud-amd64` metapackage is now installed for faster driver
recompiling on kernel upgrades.
v1 Feature **M121 release**

The M121 release of Vertex AI Workbench user-managed notebooks includes the following:

- Updated Nvidia drivers to 550.54.15 to fix an issue where Nvidia drivers failed to install on startup after Debian 11 images upgraded kernel to `linux-image-5.10.0-29-cloud-amd64`.
- The `linux-headers-cloud-amd64` metapackage is now installed for faster driver recompiling on kernel upgrades.
- TensorFlow 2.6 CPU and GPU images are deprecated. There will be no further updates to these images in future releases.
Feature The M121 release of Vertex AI Workbench managed notebooks includes the following:

- Updated the R CPU kernel from R 4.3 to R 4.4.

## May 14, 2024

**Generative AI on Vertex AI** v1 Announcement **Gemini 1.5 Flash (Preview)**

[Gemini 1.5 Flash](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/send-multimodal-prompts) ([`gemini-1.5-flash-preview-0514`](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/model-versioning#preview-models)) is available in [Preview](https://cloud.google.com/products?e=13802955&hl=en#product-launch-stages). Gemini 1.5 Flash is a multimodal model designed for fast, high volume, cost-effective text generation and chat applications. It can analyze text, code, audio, PDF, video, and video with audio.
Announcement **PaliGemma model**

The [PaliGemma](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/open-models/use-gemma) model is available. PaliGemma is a lightweight open model that's part of the Google Gemma model family. It's the Gemma model family's best model option for image captioning tasks and visual question and answering tasks. Gemma models are based on Gemini models and intended to be extended by customers.
Feature **Grounding Gemini with Google Search is GA**

The Gemini API Grounding with Google Search feature is available in [GA](https://cloud.google.com/products#product-launch-stages). This is available for Gemini 1.0 Pro models. To learn more about model grounding, see [Grounding with Google Search](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/ground-gemini#web-ground-gemini).
Feature **Batch prediction support for Gemini**

Batch prediction is available for Gemini in [preview](https://cloud.google.com/products#product-launch-stages). Available Gemini models include Gemini 1.0 Pro, Gemini 1.5 Pro, and Gemini 1.5 Flash. To get started with batch prediction, see [Get batch predictions for Gemini](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/batch-prediction-gemini).
Feature **New stable text embedding models**

The following text embedding models are available [GA](https://cloud.google.com/products#product-launch-stages):

- `text-embedding-004`
- `text-multilingual-embedding-002`

For details on how to use these models, see [Get text embeddings](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/embeddings/get-text-embeddings).
**Vertex AI** v1 Change For Ray on Vertex AI, Ray version 2.4 is no longer supported. Migrate your code to support Ray 2.9.3 or later and then delete Ray clusters that are running 2.4.
Announcement Ray on Vertex AI is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)) and includes the following updates:

- Ray version 2.9.3 and Python 3.10 are supported. For information about Ray image support policies, see [Supported versions](https://docs.cloud.google.com/vertex-ai/docs/supported-frameworks-list#ray).
- VPC peering connection is no longer required if you use public endpoints.
- Custom images are supported with Ray on Vertex AI.
- You can use custom service accounts with Ray on Vertex AI.
- A Colab template is not automatically created when you create a Ray Cluster. Instead, you can connect directly to Ray on Vertex AI clusters from [Colab Enterprise's side panel](https://docs.cloud.google.com/vertex-ai/docs/open-source/ray-on-vertex-ai/develop-application).

## April 30, 2024

**Vertex AI** v1 Feature Vertex AI custom training supports TPU v5e. For details, see [Training with TPU accelerators](https://docs.cloud.google.com/vertex-ai/docs/training/training-with-tpu-vm).

## April 29, 2024

**Agent Platform Workbench** v1 Feature **M120 release**

The M120 release of Vertex AI Workbench managed notebooks includes the following:

- Minor bug fixes for the `libcurl` package.

## April 25, 2024

**Agent Platform Workbench** v2 Feature **M120 release**

The M120 release of Vertex AI Workbench instances includes the following:

- Minor bug fixes for the `libcurl` package.
Change

### M120 Release

Fixed Minor bug fixes for the `libcurl` package.
v1 Feature **M120 release**

The M120 release of Vertex AI Workbench user-managed notebooks includes the following:

- Upgraded TensorFlow 2.15 user-managed notebooks to TensorFlow 2.15.1.
- Minor bug fixes for the `libcurl` package.

## April 18, 2024

**Generative AI on Vertex AI** v1 Feature Meta's open weight [Llama 3 model](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/open-models/use-llama) is available in the Vertex AI Model Garden.

## April 15, 2024

**Vertex AI** v1 Announcement [Persistent resource](https://docs.cloud.google.com/vertex-ai/docs/training/persistent-resource-overview) for Vertex AI custom training is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).
Feature **Vertex AI Feature Store**

The following features of Vertex AI Feature Store are now generally available ([GA](https://cloud.google.com/products#product-launch-stages)):

- **Optimized online serving** : Serve features at ultra-low latencies. For more information, see [Optimized online serving](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/online-serving-types#optimized_serving).

- **Search using embeddings** : Perform vector similarity searches to retrieve semantically similar or related features for real-time serving. You can search using embeddings if your online store is configured to support embeddings. For more information, see [Search using embeddings](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/embeddings-search).

- **Feature view sync** : Refresh or synchronize the feature data in a feature view within an online store from the feature data source in BigQuery. For more information, see [Sync feature data to online store](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/sync-data).

## April 11, 2024

**Generative AI on Vertex AI** v1 Announcement **Anthropic Claude 3.0 Opus model**

The [Anthropic Claude 3.0 Opus](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/use-claude#claude-opus) model is available in [Preview](https://cloud.google.com/products#product-launch-stages). The Claude 3.0 Opus model is an Anthropic partner model that you can use with Vertex AI. It's the most capable of the Anthropic models at performing complex tasks quickly. To learn more, see its [model card in Model Garden](https://console.cloud.google.com/vertex-ai/publishers/anthropic/model-garden/claude-3-opus).

## April 09, 2024

**Generative AI on Vertex AI** v1 Change **Change in Imagen image generation version 006 (`imagegeneration@006`) `seed` field behavior**

For the new Imagen image generation model version 006 (`imagegeneration@006`) the `seed` field behavior has changed. For the v.006 model a digital watermark is enabled by default for image generation. To be able to use a `seed` value to get deterministic output you must disable digital watermark generation by setting the following `parameter`: `"addWatermark": false`.

For more information, see the [Imagen for image generation and editing API reference](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/image-generation).
Feature **New Imagen on Vertex AI image editing model and features**

The 006 version of the Imagen 2 image editing model (`imagegeneration@006`) is now available. This model offers the following additional features:

- [Inpainting](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/edit-inpainting) - Add or remove content from a masked area of an image
- [Outpainting](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/edit-outpainting.md.txt) - Expand a masked area of an image
- [Product image editing](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/edit-product-image) - Identify and maintain a primary product while changing the background or product position

For more information, see [Model versions](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/model-versioning).
Feature **New Imagen on Vertex AI image generation model and features**

The 006 version of the Imagen 2 image generation model (`imagegeneration@006`) is now available. This model offers the following additional features:

- Additional aspect ratios (1:1, 3:4, 4:3, 9:16, 16:9)
- Digital watermark ([SynthID](https://deepmind.google/technologies/synthid/)) enabled by default
- Watermark verification\*
- New user-configurable safety features (safety setting, person/face setting)

For more information, see [Model versions](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/model-versioning) and [Generate images using text prompts](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/image/generate-images.md.txt).

\* The `seed` field can't be used while digital watermark is enabled.
Feature **Grounding Gemini and Grounding with Google Search**

The Gemini API now supports Grounding with Google Search in [Preview](https://cloud.google.com/products#product-launch-stages). Currently available for Gemini 1.0 Pro models.
Feature **Online Evaluation Service**

[Generative AI evaluation](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/evaluation-overview) supports [online evaluation](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/online-pipeline-services) in addition to [pipeline evaluation](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/side-by-side-eval). The list of supported evaluation metrics has also expanded. See [API reference](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-reference/evaluation) and [SDK reference](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/online-pipeline-services).
Feature **Supervised Tuning for Gemini**

[Supervised tuning](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/tune-gemini-overview) is available for the `gemini-1.0-pro-002 model`.
Change **Generative AI Knowledge Base**

The [Jump Start Solution: Generative AI Knowledge Base](https://console.cloud.google.com/products/solutions/details/generative-ai-knowledge-base) demonstrates how to build a simple chatbot with business- and domain-specific knowledge.
Feature **New text embedding models**

The following text embedding models are now in [Preview](https://cloud.google.com/products#product-launch-stages).

- `text-embedding-preview-0409`
- `text-multilingual-embedding-preview-0409`

When evaluated using the [MTEB benchmarks](https://huggingface.co/spaces/mteb/leaderboard), these models produce better embeddings compared to previous versions. The new models also offer [dynamic embedding sizes](https://arxiv.org/pdf/2205.13147.pdf), which you can use to output smaller embedding dimensions, with minor performance loss, to save on computing and storage costs.

For details on how to use these models, refer to the [public documentation](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/embeddings/get-text-embeddings#latest_models) and try out our [Colab](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/generative_ai/text_embedding_new_api.ipynb).
Feature **System instructions**

[System instructions](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/send-chat-prompts-gemini#system-instructions) are supported in [Preview](https://cloud.google.com/products#product-launch-stages) by the Gemini 1.0 Pro (stable version `gemini-1.0-pro-002` only) and Gemini 1.5 Pro ([Preview](https://cloud.google.com/products#product-launch-stages)) multimodal models. Use system instructions to guide model behavior based on your specific needs and use cases. For more information, see [System instructions examples](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/send-chat-prompts-gemini#system_instructions_examples).
Change **Regional APIs**

- Regional APIs are [available in 11 new countries](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/locations#available-regions) for Gemini, Imagen, and embeddings.
- [US and EU](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/locations) have machine-learning processing boundaries for the `gemini-1.0-pro-001`, `gemini-1.0-pro-002`, `gemini-1.0-pro-vision-001`, and `imagegeneration@005` models.
Feature **Text translation**

[Translate text in Vertex AI Studio](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/translate/translate-text) is available in [Preview](https://cloud.google.com/products#product-launch-stages).
Announcement **Gemini 1.0 Pro stable version 002**

The 002 version of the Gemini 1.0 Pro multimodal model (`gemini-1.0-pro-002`) is available. For more information about stable versions of Gemini models, see [Gemini model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/model-versioning#stable-version).
Feature **Generative AI on Vertex AI security control update**

[Security controls](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/security-controls) are available for the online prediction feature for Gemini 1.0 Pro and Gemini 1.0 Pro Vision.
Change **Vertex AI Studio features and updates**

- The Vertex AI Studio supports side-by-side comparison to allow users to compare up to 3 prompts in a side-by-side view.
- The Vertex AI Studio supports rapid evaluation in console and the ability to upload a ground truth response (or a model response to try to emulate).

To learn more, see [Try your prompts in Vertex AI Studio](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/start/quickstarts/quickstart)
Announcement **CodeGemma model**

The [CodeGemma](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/open-models/use-gemma) model is available. CodeGemma is a lightweight open model that's part of the Google Gemma model family. CodeGemma is the Gemma model family's code generation and code completion offering. Gemma models are based on Gemini models and intended to be extended by customers.
Announcement **Gemini 1.5 Pro (Preview)**

[Gemini 1.5 Pro](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/multimodal/send-multimodal-prompts) ([`gemini-1.5-pro-preview-0409`](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/model-versioning#preview-models)) is available in [Preview](https://cloud.google.com/products#product-launch-stages). Gemini 1.5 Pro is a multimodal model that analyzes text, code, audio, PDF, video, and video with audio.

## April 02, 2024

**Generative AI on Vertex AI** v1 Feature [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden) supports all [Text Generation Inference](https://huggingface.co/docs/text-generation-inference/en/index) supported models in [HuggingFace](https://huggingface.co/):

- Verified deployment settings for about 400 Hugging Face text generation models (including [google/gemma-7b-it](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/gemma-7b-it;action=deploy;hfSource=true), [meta-llama/Llama-2-7b-chat-hf](https://console.cloud.google.com/vertex-ai/publishers/meta-llama/model-garden/Llama-2-7b-chat-hf;action=deploy;hfSource=true), and [mistralai/Mistral-7B-v0.1](https://console.cloud.google.com/vertex-ai/publishers/mistralai/model-garden/Mistral-7B-v0.1;action=deploy;hfSource=true)).
- Other Hugging Face text generation models have unverified deployment settings that are auto generated.

## March 29, 2024

**Agent Platform Workbench** v1 Fixed **M119 release**

The M119 release of Vertex AI Workbench user-managed notebooks includes the following:

- Fixed an issue wherein Dataproc extensions caused JupyterLab to crash when remote kernels weren't available.
**Generative AI on Vertex AI** v1 Change The MedLM-large model infrastructure has been upgraded to improve
latency and stability. Responses from the model might be slightly different.

## March 18, 2024

**Agent Platform Workbench** v2 Feature **M118 release**

The M118 release of Vertex AI Workbench instances includes the following:

- Updated Nvidia drivers to R535.
Change

### M118 Release

Change Updated Nvidia drivers to R535.
v1 Feature **M118 release**

The M118 release of Vertex AI Workbench user-managed notebooks includes the following:

- PyTorch 2.1.0 with CUDA 12.1 and Python 3.10 user-managed notebooks instances are now available.
- PyTorch 2.2.0 with CUDA 12.1 and Python 3.10 user-managed notebooks instances are now available.
- Updated Nvidia drivers of older user-managed notebooks images to R535.
Feature The M118 release of Vertex AI Workbench managed notebooks includes the following:

- Updated Nvidia drivers to R535, which fixed a bug where the latest PyTorch 2.0 kernel didn't work due to outdated drivers.
**Vertex AI** v1 Change **[Vector Search](https://cloud.google.com/vertex-ai/docs/vector-search/overview) heuristics-based compaction**

Vector Search uses heuristics-based metrics assess whether to trigger compaction. This prevents unnecessary compaction, and thus reduces cost. For general information about compaction, see [Compaction](https://docs.cloud.google.com/vertex-ai/docs/vector-search/update-rebuild-index#index-compaction).

## March 08, 2024

**Vertex AI** v1 Feature **Vertex AI Feature Store**

The following features of Vertex AI Feature Store are now available in [Preview](https://cloud.google.com/products#product-launch-stages):

- **Integration of Vertex AI Feature Store with Dataplex:** Online store instances, feature views, and feature groups are now automatically registered as data assets in Data Catalog, a Dataplex feature that catalogs metadata from these resources. You can use the metadata search capability of Dataplex to search for and view the metadata of these resources. For more information, see [Search for resource metadata in Data Catalog](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/search-feature-metadata).

- **Service account configuration for feature views:** You can configure a feature view to use a dedicated service account. By default, every feature view uses the service account configured for your project. For more information, see [Configure the service account for a feature view](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/create-featureview#serviceaccount).

- **Multiple entity IDs for a feature view:** While creating or updating a feature view, you can specify multiple entity ID columns. For more information, see [Create a feature view](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/create-featureview#create_featureview_bq_embedding).

## March 05, 2024

**Vertex AI** v1 Feature **Create an empty index with Vector Search**

You can create an empty index in Vector Search for batch and for streaming. No embedding data is required at index creation time, which enables faster startup time. To learn more, see [Manage indexes](https://cloud.google.com/vertex-ai/docs/vector-search/create-manage-index).

## March 04, 2024

**Vertex AI** v1 Feature **Vertex AI Prediction**

You can now use [A3 machine types](https://docs.cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) to serve predictions.

## February 29, 2024

**Vertex AI** v1 Feature **[Vector Search](https://cloud.google.com/vertex-ai/docs/vector-search/overview) feature launch**

**Update streaming index metadata** : With this launch, you can directly update restricts and numeric restricts of data points inside `StreamUpdate` indexes without the compaction cost of a full update. To learn more, see [Update dynamic metadata](https://cloud.google.com/vertex-ai/docs/vector-search/update-rebuild-index#update-metadata).

## February 28, 2024

**Agent Platform Workbench** v2 Feature **M117 release**

The M117 release of Vertex AI Workbench instances includes the following:

- Removed the Cloud Storage browser in the left side pane in favor of the existing **Mount shared storage** button.
Change

### M117 Release

Change Removed the Cloud Storage browser in the left side pane in favor of the existing
**Mount shared storage** button.

## February 27, 2024

**Colab Enterprise** Feature VPC Service Controls has [general availability](https://cloud.google.com/products#product-launch-stages) support in Colab Enterprise.

For more information, see [Use VPC Service Controls](https://docs.cloud.google.com/colab/docs/service-controls.md.txt).

## February 26, 2024

**Vertex AI** v1 Feature Structured logging support for Vertex AI custom training. For details, see [Write code to return container logs](https://docs.cloud.google.com/vertex-ai/docs/training/code-requirements#write_code_to_return_container_logs).
Feature **Ground Multimodal Models**

<br />

Model grounding for `gemini-pro` is available in [Preview](https://cloud.google.com/products#product-launch-stages). Use grounding to
connect the `gemini-pro` model to unstructured text data stores in Vertex AI Search. Grounding lets models access and use the information in the data repositories to generate more enhanced and nuanced responses.
For more information, see [Ground multimodal models](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/multimodal/ground-gemini).

## February 21, 2024

**Vertex AI** v1 Feature **Gemma open models are available**

Gemma models, a family of lightweight, open models built from the same research and technology used to create the Gemini models, are available to run on your hardware, mobile devices, or hosted services. To learn more, see [Use Gemma open models](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/open-models/use-gemma) and the [Gemma Model Garden card](https://console.cloud.google.com/vertex-ai/publishers/google/model-garden/335).

## February 15, 2024

**Vertex AI** v1 Change The Vertex AI Gemini 1.0 Pro and Gemini 1.0 Pro Vision multimodal language models are generally available ([GA](https://cloud.google.com/products#product-launch-stages)). They have also been made available in the following regions: europe-west1, europe-west2, europe-west3, europe-west4, and europe-west9.

For more information, see the following topics:

- [Overview of multimodal models](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/multimodal/overview)
- [Multimodal prompt design](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/multimodal/design-multimodal-prompts)
- [Vertex AI Gemini API reference](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/model-reference/gemini)
- [Gemini Python SDK reference](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/multimodal/sdk-for-gemini/gemini-sdk-overview)
- [Migrate from PaLM API to Gemini API](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/migrate/migrate-palm-to-gemini)

## February 09, 2024

**Vertex AI** v1 Feature **Multimodal embeddings video support is generally available**

Embeddings for video data is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)) using the multimodal embedding model (`multimodalembedding`). For more information, see the [product documentation](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/embeddings/get-multimodal-embeddings#vid-img-txt-request).

This features incurs pricing based on the mode you use. For more information,
see [pricing](https://cloud.google.com/vertex-ai/pricing#generative_ai_models).

## February 08, 2024

**Agent Platform Workbench** v1 Feature **M116 release**

The M116 release of Vertex AI Workbench user-managed notebooks includes the following:

- Updated custom container user-managed notebooks to use NVIDIA driver version 535.104.05.
- Fixed bugs in custom container user-managed notebooks where GPUs either wouldn't attach to the container properly, or detached after some time.
Feature The M116 release of Vertex AI Workbench managed notebooks includes the following:

- Fixed a bug (present in versions M113 through M115) that prevented new local kernels from being usable.

## February 07, 2024

**Vertex AI** v1 Feature The following models have been added to [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden):

- Stable Diffusion XL LCM: The Latent Consistency Model (LCM) enhances text-to-image generation in Latent Diffusion Models by enabling faster and high-quality image creation with fewer steps.
- LLaVA 1.5: Deploy LLaVA 1.5 models.
- PyTorch-ZipNeRF: The Pytorch-ZipNeRF model is a state-of-the-art implementation of the ZipNeRF algorithm in the Pytorch framework, designed for efficient and accurate 3D reconstruction from 2D images.
- LLaMA 2 (Quantized): A quantized version of Meta's Llama 2 models.
- WizardLM: WizardLM is a large language model (LLM) developed by Microsoft, fine-tuned on complex instructions by adapting the Evol-Instruct method.
- WizardCoder: WizardCoder is a large language model (LLM) developed by Microsoft, fine-tuned on complex instructions by adapting the Evol-Instruct method to the domain of code.
- AutoGluon: With AutoGluon you can train and deploy high-accuracy machine learning and deep learning models for tabular data.
- Lama (Large mask inpainting): Use Large Mask Inpainting with fast Fourier convolutions (FFCs), a high receptive field perceptual loss, and large training masks for resolution-robust image inpainting.
Change The following changes have been made to [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden):

- Added one-click tuning button, and dedicated deployment, tuning, quantization, and evaluation notebooks for Llama 2.
- Added one-click deployment button for more than 20 models with pre-trained OSS artifacts, including `Salesforce/blip-image-captioning-base` and `timbrooks/instruct-pix2pix`.
- Supported CodeLlaMA70b with notebooks and the one-click deployment button.
- Added tuning notebooks for Mistral models.
- Added serving notebooks for Stable Video Diffusion Img2Vid XT. These notebooks are used for research purposes.

## February 05, 2024

**Vertex AI** v1 Feature **Query an index from the Vector Search console**

Vector Search has launched an improved console experience for querying both private and public deployed indexes, now available in [Preview](https://cloud.google.com/products/#product-launch-stages). From the console, you can create an index and endpoint, deploy the index to the endpoint, and query the index for nearest neighbors. For more information, see [Manage indexes](https://cloud.google.com/vertex-ai/docs/vector-search/create-manage-index).

## January 29, 2024

**Vertex AI** v1 Feature **Vertex Prediction**

You can now customize more deployment parameters when uploading your models, such as shared memory allocation and custom startup and readiness probes. These parameters may be useful when deploying LLMs.

For more information, see [Deploy generative AI models](https://docs.cloud.google.com/vertex-ai/docs/predictions/deploy-genai), [Custom container requirements for prediction](https://docs.cloud.google.com/vertex-ai/docs/predictions/custom-container-requirements), and [`ModelContainerSpec`](https://docs.cloud.google.com/vertex-ai/docs/reference/rest/v1/ModelContainerSpec).

## January 19, 2024

**Agent Platform Workbench** v2 Feature **M115 release**

The M115 release of Vertex AI Workbench instances includes the following:

- Added support for `venv` kernels.
Change

### M115 Release

Feature Added support for `venv` kernels.
v1 Feature The M115 release of Vertex AI Workbench managed notebooks includes the following:

- Fixed the BigQuery connector within PySpark containers.
Feature **M115 release**

The M115 release of Vertex AI Workbench user-managed notebooks includes the following:

- Added support for TensorFlow 2.15 with Python 3.10 on Debian 11.
- Added support for TensorFlow 2.14 with Python 3.10 on Debian 11.

## January 16, 2024

**Agent Platform Workbench** v1 Deprecated Vertex AI Workbench user-managed notebooks is [deprecated](https://docs.cloud.google.com/vertex-ai/docs/deprecations.md.txt). On January 30, 2025, support for user-managed notebooks will end and the ability to create user-managed notebooks instances will be removed. Existing instances will continue to function but patches, updates, and upgrades won't be available. To continue using Vertex AI Workbench, you can [migrate your user-managed notebooks instances to Vertex AI Workbench instances](https://docs.cloud.google.com/vertex-ai/docs/workbench/user-managed/migrate-to-instances).
Deprecated Vertex AI Workbench managed notebooks is [deprecated](https://docs.cloud.google.com/vertex-ai/docs/deprecations.md.txt). On January 30, 2025, support for managed notebooks will end and the ability to create managed notebooks instances will be removed. Existing instances will continue to function but patches, updates, and upgrades won't be available. To continue using Vertex AI Workbench, you can [migrate your managed notebooks instances to Vertex AI Workbench instances](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/migrate-to-instances).

## January 12, 2024

**Vertex AI** v1 Feature Model tuning for the `textembedding-gecko` and `textembedding-gecko-multilingual`
models is available in [GA](https://cloud.google.com/products#product-launch-stages).
You can use supervised fine-tuning to tune the `textembedding-gecko` model.
For more information, see [Tune text embeddings](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/models/tune-embeddings).

## January 08, 2024

**Vertex AI** v1 Change **AutoSxS evaluates LLMs side by side**

The [automatic side-by-side (AutoSxS) evaluation
tool](https://cloud.google.com/vertex-ai/docs/generative-ai/models/side-by-side-eval)
is available in Preview to A/B test the performance of your LLMs or pre-generated
predictions. It's comparable to human evaluators, yet faster, available
on-demand, and more cost-efficient.

## January 05, 2024

**Vertex AI** v1 Change **Generative AI on Vertex AI regional expansion**

Generative AI on Vertex AI features for Batch Prediction and Model Evaluation are available in [12 additional Google Cloud regions](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/learn/locations#available-regions).

## December 29, 2023

**Vertex AI** v1 Change **Vertex AI regional expansion**

Vertex AI features for AutoML Forecasting, AutoML Tabular, Batch Prediction, Online Prediction, Pipelines, Training, Vector Search, and Vizier are available in [10 additional Google Cloud regions](https://docs.cloud.google.com/vertex-ai/docs/general/locations#feature-availability).

## December 27, 2023

**Vertex AI** v1 Change **Vertex Prediction**

Quota for *Custom model serving* is now calculated based on your [deployed model's](https://docs.cloud.google.com/vertex-ai/docs/general/deployment#scaling) real-time usage of compute resources.

Previously, compute resources, such as the number of Nvidia A100 GPUs being used, were deducted from your project's quota based on the deployed model's [`maxReplicaCount`](https://docs.cloud.google.com/vertex-ai/docs/reference/rest/v1/DedicatedResources).

This change lets you deploy models based on actual compute usage rather than max usage, but it can prevent your deployed models from autoscaling if your quota is exhausted.

## December 18, 2023

**Vertex AI** v1 Announcement **Model Garden** supports:

- [Hyperparameter tuning](https://docs.cloud.google.com/vertex-ai/docs/training/hyperparameter-tuning-overview) and customized datasets for [OpenLLaMA models](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_openllama_peft.ipynb) using the [dataset format used by supervised tuning](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/gemini-supervised-tuning#dataset-format) in Vertex AI.
- GPTQ conversions for [falcon-instruct models](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_falcon_instruct_peft.ipynb).
- [Latent Consistent Models](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_stable_diffusion_xl_lcm.ipynb), and [research purpose only SDXL-Turbo models](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_stable_diffusion_xl_1_0.ipynb) to stable diffusion XL notebooks.
- Mixtral 8x7B models in the [Mistral notebook](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/model_garden/model_garden_pytorch_mistral_deployment.ipynb).

## December 14, 2023

**Agent Platform Workbench** v1 Feature **M114 release**

The M114 release of Vertex AI Workbench user-managed notebooks includes the following:

- Starting with this release, Python 3.7 is no longer available.
- Upgraded R to 4.3 on Debian 11 Python 3.10 instances.
- Upgraded JupyterLab to 3.6.6.
Feature The M114 release of Vertex AI Workbench managed notebooks includes the following:

- Starting with this release, Python 3.7 is no longer available.
- Added new Dataproc extension for remote kernels.
- Upgraded JupyterLab to 3.6.6.
- Fixed an issue that sometimes prevented users from running or scheduling notebooks using a default kernel.
**Vertex AI** v1 Feature **Vertex AI Prediction**

You can now use [Cloud TPU v5e](https://docs.cloud.google.com/tpu/docs/v5e-inference) to serve online predictions. For more information, see [Use Cloud TPUs for online prediction](https://docs.cloud.google.com/vertex-ai/docs/predictions/use-tpu).

## December 13, 2023

**Vertex AI** v1 Feature **Vertex AI Gemini models**

Vertex AI Gemini Pro and Gemini Pro Vision multimodal language models are available in [Preview](https://cloud.google.com/products#product-launch-stages). For more information, see the following topics:

- [Overview of multimodal models](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/multimodal/overview)
- [Multimodal prompt design](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/multimodal/design-multimodal-prompts)
- [Vertex AI Gemini API reference](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/model-reference/gemini)
- [Gemini Python SDK reference](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/multimodal/sdk-for-gemini/gemini-sdk-overview)
- [Migrate from PaLM API to Gemini API](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/migrate/migrate-palm-to-gemini)
Feature **Imagen 2 General Availability**

The 005 version of Imagen's image generation model (`imagegeneration@005`) is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)) for image generation tasks. This model version is now the default for image generation tasks. For more information, see the [product documentation](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/image/generate-images).

For general information about Imagen models and versions, see [Imagen model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/image/model-versioning).

## December 12, 2023

**Vertex AI** v1 Feature **Text embedding model 003 (textembedding-gecko@003) available**

The updated stable version of the text embedding foundation model, `textembedding-gecko@003`, is available. `textembedding-gecko@003` features improved quality compared to the previous stable versions, `textembedding-gecko@001` and `textembedding-gecko@002`. For more information on model versions, see [Model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/learn/model-versioning).

## December 08, 2023

**Vertex AI** v1 Change **Generative AI on Vertex AI security control update**

The [Access Transparency (AXT) security control](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/genai-security-controls) is available for the following features:

- Embeddings for Multimodal online prediction
- Imagen on Vertex AI online prediction
- Imagen on Vertex AI tuning

## December 06, 2023

**Vertex AI** v1 Feature Version `@002` of the models for text, chat, code, and code chat are
available. The `@002` model versions include improved prompt responses.
The `@002` models are:

- `text-bison@002`
- `chat-bison@002`
- `code-bison@002`
- `codechat-bison@002`

To ensure that you always use the *stable* model version, specify the model
identifier with the version number. For example, `text-bison@002`. For more
information, see [Model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/learn/model-versioning).
Feature Version 2 of the stable version of the Codey code completion foundation model, named `code-gecko@002`, is available. `code-gecko@002` features improved quality and reduced latency compared to the previous stable version, `code-gecko@001`. These improvements can lead to a higher rate of acceptance.

## December 05, 2023

**Vertex AI** v1 Feature **Grounding with Vertex AI Search**

<br />

Model grounding is available in ([Preview](https://cloud.google.com/products#product-launch-stages)). Use grounding to
connect the `text-bison` and `chat-bison` models to unstructured data stores in Vertex AI Search.
Grounding lets models access and use the information in the data repositories to generate more enhanced and nuanced responses.
For more information, see the [Grounding Overview](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/grounding/overview).

## December 01, 2023

**Vertex AI** v1 Change The following Vertex AI [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden) updates are available:

- Updated default model deployment settings with L4 GPUs, such as LLaMA2, falcon-instruct, openllama, Stable Diffusion 1.5, 2.1, and XL models.
- Support for [hyperparameter tuning](https://docs.cloud.google.com/vertex-ai/docs/training/hyperparameter-tuning-overview) and customized datasets for LLaMA2 models using the [dataset format used by supervised tuning](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/models/tune-text-models-supervised#dataset-format) in Vertex AI.
- Recommended LoRA and QLoRA settings for large language model tuning in Vertex AI. For details, see [LoRA and QLoRA recommendations for LLMs](https://docs.cloud.google.com/vertex-ai/docs/model-garden/lora-qlora).
- Support for AWQ and GPTQ conversions for LLaMA2 and OpenLLaMA models.
- Benchmark reports for [ViT pytorch and JAX training](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/community-content/vertex_model_garden/benchmarking_reports/jax_vit_benchmarking_report.md), [OpenLLaMA 3b/7b/13b hyperparameter tuning](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/community-content/vertex_model_garden/benchmarking_reports/pytorch_openllama_hyperparameter_tuning_benchmark_report.md), and [Stable Diffusion 1.5 tuning and serving](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/community-content/vertex_model_garden/benchmarking_reports/stable_diffusion_v1-5_benchmarking_report.md).

## November 30, 2023

**Vertex AI** v1 Feature The Unicorn model size for PaLM 2 for Text is generally available ([GA](https://cloud.google.com/products/#product-launch-stages)). The `text-unicorn` model provides improved response quality and reasoning capability compared to the `text-bison` model. For details, see [Model information](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/learn/models).
Announcement Vertex AI's integration of model and dataset metadata into Dataplex's Data Catalog service is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)). Search and discover these assets across projects and regions in Dataplex. Learn more at [Use Data Catalog to search for model and dataset resources](https://docs.cloud.google.com/vertex-ai/docs/model-registry/model-data-catalog).

**Note** : For datasets of type `TEXT_PROMPT`, navigating in the UI from Data Catalog back to Vertex AI (via the **Open in Vertex AI** button, or using the **Resource URL** link) results in a blank page. This is a known issue and is expected to be fixed in the near future. To directly view `TEXT_PROMPT` datasets in Vertex AI, navigate to the [**Generative AI My Prompts**](https://console.cloud.google.com/vertex-ai/generative/language/my-prompts) tab.

## November 24, 2023

**Vertex AI** v1 Feature **ComputeToken API now available in Preview**

The ComputeToken API is now available in ([Preview](https://cloud.google.com/products#product-launch-stages)).
You can use this API to get a list of tokens for a given prompt. A token is a way to represent a common sequence of characters found in a text input. To learn more, see [Get a list of tokens](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/compute-token).

## November 17, 2023

**Vertex AI** v1 Feature **Vertex AI Feature Store**

The following features of the new and improved Vertex AI Feature Store are now generally available ([GA](https://cloud.google.com/products#product-launch-stages)):

- **Feature Registry** : Register your feature data sources in BigQuery by creating feature groups and features. For more information, see [Create a feature group](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/create-featuregroup) and [Create a feature](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/create-feature).

- **Cloud Bigtable online serving** : Serve features from one or more BigQuery data sources. You can set up Cloud Bigtable online serving by [defining online serving clusters called online store instances](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/create-onlinestore) and [creating feature views](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/create-featureview) within the online store instances.

Note that the following features of Vertex AI Feature Store are still in [Preview](https://cloud.google.com/products#product-launch-stages):

- Serve features at ultra-low latencies with [Optimized online serving](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/online-serving-types).
- [Sync data](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/sync-data) in a feature view within an online store.
- [Retrieve vector embeddings](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/embeddings-search) for real-time serving.

For more information, see [About Vertex AI Feature Store](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/overview).

## November 16, 2023

**Agent Platform Workbench** v2 Feature **M113 release**

The M113 release of Vertex AI Workbench instances includes the following:

- Added the Dataproc JupyterLab plugin to Vertex AI Workbench instances. To get started, see [Create a Dataproc-enabled instance](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/create-dataproc-enabled).
- When using an instance's Google Cloud CLI, `gcloud config` is preset with the following defaults:
  - `project` is set to your instance's project.
  - Your compute region is set to your instance's region.
  - Your Dataproc region is set to your instance's region.
- Fixed an issue that prevented Dataproc kernels from working.
- Fixed a CORS (cross-origin resource sharing) error.
Change

### M113 Release

Feature Added the Dataproc JupyterLab plugin to Vertex AI Workbench instances. To get
started, see [Create a Dataproc-enabled
instance](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/create-dataproc-enabled).
Change When using an instance's Google Cloud CLI, `gcloud config` is preset with the
following defaults: \* `project` is set to your instance's project. \* Your
compute region is set to your instance's region. \* Your Dataproc region is set
to your instance's region.
Fixed Fixed an issue that prevented Dataproc kernels from working.
Fixed Fixed a CORS (cross-origin resource sharing) error.
v1 Feature **M113 release**

The M113 release of Vertex AI Workbench user-managed notebooks includes the following:

- Miscellaneous bug fixes and improvements in Python 3.10 notebooks.

## November 13, 2023

**Vertex AI** v1 Feature **Numerical filtering available in Vertex AI Vector Search**

With [Vector Search](https://docs.cloud.google.com/vertex-ai/docs/vector-search/overview) you can restrict results by "filtering" your index results. In addition to filtering by using categorical restrictions, you can now use numeric filtering.
To learn more, see [Filter vector matches](https://docs.cloud.google.com/vertex-ai/docs/vector-search/filtering).

## November 10, 2023

**Vertex AI** v1 Feature **Generative AI on Vertex AI**

<br />


[Security controls](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/genai-security-controls) are available for additional Generative AI on Vertex AI features.

## November 07, 2023

**Vertex AI** v1 Announcement [Training on TPU VMs](https://docs.cloud.google.com/vertex-ai/docs/training/training-with-tpu-vm) is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

## November 03, 2023

**Vertex AI** v1 Announcement **New `textembedding-gecko` and `textembedding-gecko-multilingual` stable model versions**

The following stable model versions are available in Generative AI on Vertex AI:

- `textembedding-gecko@002`
- `textembedding-gecko-multilingual@001`

For more information on model versions, see [Model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/learn/model-versioning).
Feature The following models have been added to [Model Garden](https://console.cloud.google.com/vertex-ai/model-garden):

- ImageBind: Multimodal embedding model.
- Vicuna v1.5: LLM finetuned based on llama2.
- OWL-ViT v2: SoTA Open Vocabulary Object Detection model.
- DITO: SoTA Open Vocabulary Object Detection model.
- NLLB: Multi-language translation model.
- Mistral-7B: SoTA LLM at small size.
- BioGPT: LLM finetuned for biomedical domain.
- BiomedCILP: Multimodal foundational model finetuned for biomedical domain.

To see a list of all available models, see [Explore models in Model Garden](https://docs.cloud.google.com/vertex-ai/docs/start/explore-models).
Change **Model Garden**

- Improved language model serving throughput. For details, see [Serving open source large language models efficiently on Model Garden](https://cloud.google.com/blog/products/ai-machine-learning/serve-open-source-llms-on-google-cloud). Notebooks in the relevant model cards have been updated accordingly.
- Inference speed up to 2 times faster compared with original implementation for Stable Diffusion 1.5, 2.1, and XL models.
- Improved the workflow of the **Deploy** button in all supported model cards.
- Updated notebooks for Llama2, OpenLlama, and Falcon Instruct with suggested machine specs for model serving, and [EleutherAI's evaluation harness](https://github.com/EleutherAI/lm-evaluation-harness) dockers for model evaluation.

## November 02, 2023

**Vertex AI** v1 Change **Generative AI support on Vertex AI**

Generative AI on Vertex AI can be accessed through [12 regional APIs](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/learn/locations-genai) in North America, Europe, and Asia. Regional APIs let customers control where data is stored at-rest.

## October 30, 2023

**Vertex AI** v1 Security [Deep Learning VM Images](https://docs.cloud.google.com/deep-learning-vm/docs/images.md.txt) is a set of prepackaged virtual machine images with a deep learning framework that are ready to be run out of the box. Recently, an out-of-bounds write vulnerability was discovered in the `ReadHuffmanCodes()` function in the `libwebp` library. This might impact images that use this library.

Google Cloud continuously scans its publicly published images and updates the packages to assure patched distros are included in the latest releases available for customer adoption. Deep Learning VM Images have been updated to ensure that the latest VM images include the patched distros. Customers adopting the latest VM images are not exposed to this vulnerability.

For more information, see the [Vertex AI security bulletin](https://docs.cloud.google.com/vertex-ai/docs/security-bulletins#GCP-2023-036).

## October 17, 2023

**Vertex AI** v1 Announcement **Vertex AI Vector Search Improvements**

Vector Search has improved the initial index creation process for smaller indexes (\<100MB), reducing time to build from about 1 hour to about 5 mins. To get started, see [Vector Search quickstart](https://docs.cloud.google.com/vertex-ai/docs/vector-search/quickstart) to create an index.
Feature **New Vertex AI Vector Search Console**

Vector Search has launched a console experience in Google Cloud for creating and deploying indexes, now available in [Preview](https://cloud.google.com/products/#product-launch-stages). From the console, you can create indexes, and create public or VPC endpoints for your indexes, and deploy. For more information, see [Manage indexes](https://docs.cloud.google.com/vertex-ai/docs/vector-search/create-manage-index).

## October 11, 2023

**Colab Enterprise** Feature Colab Enterprise is now generally available ([GA](https://cloud.google.com/products/#product-launch-stages)). Colab Enterprise combines the popular collaborative features of Colaboratory with the security and compliance capabilities of Google Cloud. Colab Enterprise includes:

- Sharing and collaborating functionality, with IAM access control.
- Google-managed compute and runtime provisioning, with configurable [runtime templates](https://docs.cloud.google.com/colab/docs/create-runtime-template.md.txt).
- Integrations with Vertex AI and [BigQuery](https://docs.cloud.google.com/bigquery/docs/explore-data-colab.md.txt).
- Inline code completion with [Duet AI](https://docs.cloud.google.com/duet-ai/docs/overview) ([Preview](https://cloud.google.com/products/#product-launch-stages)) assistance.
- End-user credential authentication for running your notebook code.
- [Idle shutdown](https://docs.cloud.google.com/colab/docs/idle-shutdown.md.txt) for runtimes ([Experimental](https://cloud.google.com/products/#product-launch-stages)).

To get started, see [Introduction to Colab Enterprise](https://docs.cloud.google.com/colab/docs/introduction.md.txt) or [create a notebook](https://docs.cloud.google.com/colab/docs/create-console-quickstart.md.txt) and start coding.

## October 10, 2023

**Agent Platform Workbench** v1 Feature **M112 release**

The M112 release of Vertex AI Workbench user-managed notebooks includes the following:

- Miscellaneous bug fixes and improvements.

## October 05, 2023

**Vertex AI** v1 Feature **Ray on Vertex AI is now available in Preview**

[Ray](https://docs.ray.io/en/latest/ray-overview/index.html) is an open-source framework for scaling AI and
Python applications. Ray provides the infrastructure to perform distributed
computing and parallel processing for your machine learning workflow.

You can now create Ray clusters and develop your Ray applications on Vertex AI. This feature is in [Preview](https://cloud.google.com/products#product-launch-stages). For more information, see [Ray on Vertex AI overview](https://docs.cloud.google.com/vertex-ai/docs/open-source/ray-on-vertex-ai/overview).

## October 04, 2023

**Vertex AI** v1 Feature **Model tuning for the `textembedding-gecko` model is now available in Preview**

You can now use supervised fine-tuning to tune the `textembedding-gecko` model. This feature is in ([Preview](https://cloud.google.com/products#product-launch-stages)).
For more information, see [Tune text embeddings](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/models/tune-embeddings).
Feature **Vertex AI Feature Store**

The new and improved Vertex AI Feature Store is now available in [Preview](https://cloud.google.com/products#product-launch-stages). With the new Vertex AI Feature Store you can streamline your feature management in the following ways:

- Store and maintain your offline feature data in BigQuery, taking advantage of the data management capabilities of BigQuery. In the new Vertex AI Feature Store, BigQuery serves as the offline store. You don't need to copy or import feature data to an offline store in Vertex AI.

- Register your feature data sources in BigQuery by creating feature groups and features.

- Define online serving clusters called online store instances; and then serve features from one or more BigQuery data sources, by aggregating them in a feature view within an online store instance. Use Optimized online serving for ultra-low latency needs and Cloud Bigtable online serving for high data volumes.

- Retrieve vector embeddings stored in BigQuery for real-time serving.

For more information, see [About Vertex AI Feature Store](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/overview).
Feature **Vertex AI Prediction**

You can now use [C3 machine types](https://docs.cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) to serve predictions.

## October 03, 2023

**Vertex AI** v1 Security TorchServe is used to host PyTorch machine learning models for online prediction. Vertex AI provides pre-built PyTorch model serving containers which depend on TorchServe. Vulnerabilities were recently discovered in TorchServe which would allow an attacker to take control of a TorchServe deployment if its model management API is exposed. Customers with PyTorch models deployed to Vertex AI online prediction are not affected by these vulnerabilities, since Vertex AI does not expose TorchServe's model management API. Customers using TorchServe outside of Vertex AI should take precautions to ensure their deployments are set up securely.

For more information, see the [Vertex AI security bulletin](https://docs.cloud.google.com/vertex-ai/docs/security-bulletins#gcp-2023-029).

## September 25, 2023

**Agent Platform Workbench** v2 Feature Vertex AI Workbench instances are now generally available ([GA](https://cloud.google.com/products/#product-launch-stages)). Vertex AI Workbench instances combine features from managed notebooks and user-managed notebooks to provide a robust data science solution. Supported features include:

- Idle timeout
- BigQuery and Cloud Storage integrations
- End-user and service account authentication
- VPC Service Controls
- Customer managed encryption keys (CMEK) and Cloud External Key Manager (Cloud EKM)
- Health status monitoring
- Scheduled notebook runs
- Dataproc integration

To get started, see [Introduction to Vertex AI Workbench instances](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/introduction).

## September 18, 2023

**Agent Platform Workbench** v1 Deprecated Debian 10 and Python 3.7 images have reached their end of patch and support life for Vertex AI Workbench managed notebooks and user-managed notebooks. Debian 11 and Python 3.10 images are available.

## September 14, 2023

**Agent Platform Workbench** v1 \& v2 Feature **M111 release**

The M111 release of Vertex AI Workbench instances includes the following:

- Miscellaneous software updates.
Feature The M111 release of Vertex AI Workbench user-managed notebooks includes the following:

- PyTorch 2.0 user-managed notebooks instances now include PyTorch XLA 2.0.
- Miscellaneous software updates.
Feature The M111 release of Vertex AI Workbench managed notebooks includes the following:

- Miscellaneous software updates.
v2 Change

### M111 Release

Change Miscellaneous software updates.

## September 08, 2023

**Vertex AI** v1 Feature **Vertex AI Prediction**

You can now use [A2 Ultra machines](https://docs.cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) to serve predictions in `us-central1`, `us-east4`, `europe-west4`, and `asia-southeast1`. Each A2 Ultra machine has a fixed number of NVIDIA A100 80GB GPUs attached.

## September 06, 2023

**Vertex AI** v1 Change **Vertex AI Prediction**

The following prebuilt containers for prediction have been updated:

- tf2-cpu.2-12
- tf2-gpu.2-12
- tf2-cpu.2-11
- tf2-gpu.2-11
- tf2-cpu.2-10
- tf2-gpu.2-10
- tf2-cpu.2-9
- tf2-gpu.2-9
- tf2-cpu.2-8
- tf2-gpu.2-8
- sklearn-cpu.1-2
- xgboost-cpu.1-7
- pytorch-cpu.2-0
- pytorch-gpu.2-0
- pytorch-cpu.1-13
- pytorch-gpu.1-13

To update your containers, redeploy your models. To learn more, see [Vertex AI framework support policy](https://docs.cloud.google.com/vertex-ai/docs/framework-support-policy) and [Prebuilt containers for prediction](https://docs.cloud.google.com/vertex-ai/docs/predictions/pre-built-containers).

## September 01, 2023

**Vertex AI** v1 Change **Pricing update**

The pricing for `text-bison` has been reduced to $0.0005 per 1,000 input and output characters. For details, see [Vertex AI Pricing](https://cloud.google.com/vertex-ai/pricing).

## August 31, 2023

**Vertex AI** v1 Feature Experiment management: The Google Cloud console supports visualization of your model's performance changes over steps during training, and shows advanced run comparisons. To learn more, see [Compare and analyze runs: Google Cloud console](https://docs.cloud.google.com/vertex-ai/docs/experiments/compare-analyze-runs#console-compare-analyze-runs).

## August 29, 2023

**Colab Enterprise** Feature Colab Enterprise is now available in [Preview](https://cloud.google.com/products/#product-launch-stages). Colab Enterprise combines the popular collaborative features of Colaboratory with the security and compliance capabilities of Google Cloud. Colab Enterprise includes:

- Sharing and collaborating functionality, with IAM access control.
- Google-managed compute and runtime provisioning, with configurable runtime templates.
- Integrations with Vertex AI and BigQuery.
- Inline code completion with [Duet AI](https://docs.cloud.google.com/duet-ai/docs/overview) assistance.
- End-user credential authentication for running your notebook code.

To get started, see [Introduction to Colab Enterprise](https://docs.cloud.google.com/colab/docs/introduction.md.txt) or [create a notebook](https://docs.cloud.google.com/colab/docs/create-console-quickstart.md.txt) and start coding.
**Vertex AI** v1 Feature **Model tuning for the `chat-bison` model is now available in Preview**

You can now use supervised fine-tuning to tune the `chat-bison` model. This feature is in ([Preview](https://cloud.google.com/products#product-launch-stages)).
For more information, see [Tune text models](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/models/tune-code-models).
Feature **New embedding model now available in Preview**

Generative AI support on Vertex AI users can now create embeddings using a new model trained on a wide range of non-English languages in ([Preview](https://cloud.google.com/products#product-launch-stages)).

- `textembedding-gecko-multilingual`

To learn more, see [Get text embeddings](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/embeddings/get-text-embeddings).
Feature **New Generative AI support on Vertex AI models and expanded language support**

Generative AI support on Vertex AI has been updated to include new language
model candidates (*latest* models), language models that support input and
output tokens up to 32k, and more supported languages.

For details, see [Available models](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/learn/models)
and [Model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/learn/model-versioning).
Feature **Model tuning for the `text-bison` model is now Generally Available (GA)**

Tuning the `text-bison` model with supervised fine-tuning (SFT) is now Generally Available [(GA)](https://cloud.google.com/products#product-launch-stages) .
For more information, see [Tune text models](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/models/tune-text-models).
Feature **Stream responses from Generative AI models**

Generative AI model streaming support is now Generally Available [(GA)](https://cloud.google.com/products#product-launch-stages). After you send a prompt, the model returns response tokens as they're generated instead of waiting for the entire output to be available.

Supported models are:

- `text-bison`
- `chat-bison`
- `code-bison`
- `codechat-bison`

To learn more, see [Stream responses from Generative AI models](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/learn/streaming).
Feature **Imagen on Vertex AI now offers the following [Generally Available (GA)](https://cloud.google.com/products#product-launch-stages) features:**

- [Subject model tuning (standard)](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/image/fine-tune-model)\*
- [Style model tuning](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/image/fine-tune-style)\*

\* Restricted access feature.

For more information about Imagen or how to get access to restricted GA features, see the
[Imagen on Vertex AI overview](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/image/overview).
Feature **Metrics-based model evaluation**

You can evaluate the performance of foundation models and tuned models against an evaluation dataset for classification, summarization, question answering, and general text generation. This feature is available in ([Preview](https://cloud.google.com/products#product-launch-stages)).

To learn more, see [Evaluate model performance](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/models/evaluate-models).
Feature **Vertex AI Codey APIs language support**

Vertex AI Codey APIs now support additional programming languages. For more information, see [Supported coding languages](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/code/code-models-overview#supported_coding_languages)
Announcement **Vertex AI Vector Search** is the new product name for *Vertex AI Matching Engine*.
Feature **Vertex AI Codey APIs now support supervised fine-tuning (SFT)**

The code chat (`codechat-bison`) and code generation (`code-bison`) Vertex AI Codey APIs models now support supervised fine-tuning (SFT). The supervised-fine tuning for Vertex AI Codey APIs models feature is in ([Preview](https://cloud.google.com/products#product-launch-stages)). For more information, see [Tune code models](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/models/tune-code-models).
Feature **Reinforcement learning from human feedback (RLHF) tuning for `text-bison`**
The Generative AI text generation foundation model (`text-bison`) now supports RLHF tuning. The RLHF tuning feature is in ([Preview](https://cloud.google.com/products#product-launch-stages)).
For more information, see [Use RLHF model tuning](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/models/tune-models#genai-rlhf-tuning).
Feature The Vertex AI Pipelines **Template Gallery** is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)). The Template Gallery contains Google-authored pipeline and component templates to bootstrap your MLOps practice. Customize and run the templates as-is or embed them into your own pipelines. For more information, see [Use a prebuilt template from the Template Gallery](https://docs.cloud.google.com/vertex-ai/docs/pipelines/use-template-gallery).
Feature Vertex AI Model Registry Models and managed datasets are now synced to Dataplex's Data Catalog service. [Data Catalog](https://docs.cloud.google.com/data-catalog/docs/concepts/overview.md.txt) enables organization-wide search and discovery of data artifacts, while still maintaining IAM boundaries. The sync and search of these assets is available in [Preview](https://cloud.google.com/products/#product-launch-stages). For more information, see [Data Catalog documentation](https://docs.cloud.google.com/data-catalog/docs/how-to/search.md.txt).
Feature **CountToken API now available in Preview**

The CountToken API is now available in ([Preview](https://cloud.google.com/products#product-launch-stages)). You can use this API to get the token count and the number of billable characters for a prompt. To learn more, see [Get token count](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/get-token-count).

## August 28, 2023

**Vertex AI** v1 Feature Tabular Workflow for Forecasting is available in [Preview](https://cloud.google.com/products#product-launch-stages). For documentation, refer to [Tabular Workflow for Forecasting](https://docs.cloud.google.com/vertex-ai/docs/tabular-data/tabular-workflows/forecasting).

## August 22, 2023

**Vertex AI** v1 Feature Vertex AI custom training has launched [persistent resources](https://docs.cloud.google.com/vertex-ai/docs/training/persistent-resource-overview) in [Preview](https://cloud.google.com/products#product-launch-stages). A persistent resource is a long-running cluster of machines that you can use to run custom training jobs. Once created, the persistent resource remains available for future training jobs, so you don't have to wait for compute resources to be provisioned each time you want to train a model.

## August 18, 2023

**Vertex AI** v1 Feature The Vertex AI Matching Engine public endpoint is now generally available ([GA](https://cloud.google.com/products/#product-launch-stages)). For information about how to get started, see [Matching Engine Setup](https://docs.cloud.google.com/vertex-ai/docs/matching-engine/match-eng-setup/match-eng-setup).

## August 11, 2023

**Vertex AI** v1 Change Generative AI on Vertex AI supports CMEK, VPC Service Controls, Data Residency, and Access Transparency. For more information, see [Security controls](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/genai-security-controls).

## August 10, 2023

**Agent Platform Workbench** v1 \& v2 Feature **M110 release**

The M110 release of Vertex AI Workbench user-managed notebooks includes the following:

- Added support for TensorFlow 2.13 with Python 3.10 on Debian 11.
- Added support for TensorFlow 2.8 with Python 3.10 on Debian 11.
- Miscellaneous software updates.
Deprecated TensorFlow 2.9 user-managed instances are deprecated.
Feature The M110 release of Vertex AI Workbench managed notebooks includes the following:

- Increased shared memory size to available memory capacity.
- Added support for Python 3.10 on Debian 11.
  - Added support for PyTorch 2.0 with Python 3.10.

> [!NOTE]
> **Note:** Python 3.7 on Debian 10 images are still available.

## August 09, 2023

**Vertex AI** v1 Feature **Imagen Multimodal embeddings available in GA**

Imagen on Vertex AI now offers the following [GA](https://cloud.google.com/products/#product-launch-stages) feature:

- Multimodal embeddings

This feature incurs different [pricing](https://cloud.google.com/vertex-ai/pricing#generative_ai_models)
based on if you use image input or text input. For more information, see the
[multimodal embeddings](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/embeddings/get-image-embeddings)
feature page.

## August 02, 2023

**Vertex AI** v1 Change Updated prebuilt images for [Tensorflow 2.11](https://docs.cloud.google.com/vertex-ai/docs/training/pre-built-containers#tensorflow) are now available.
Change Prebuilt containers to perform custom training with [TensorFlow 2.12](https://docs.cloud.google.com/vertex-ai/docs/training/pre-built-containers#tensorflow) are now generally available [(GA)](https://cloud.google.com/products#product-launch-stages).

## August 01, 2023

**Vertex AI** v1 Feature The schedules API for Vertex AI Pipelines is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)). You can schedule recurring pipeline runs in Vertex AI by specifying a frequency in [cron](https://en.wikipedia.org/wiki/Cron) syntax, and optionally the start time and/or end time. Additionally, you can pause, resume, update, and delete schedules.For more information, see [Schedule a pipeline run with scheduler API](https://docs.cloud.google.com/vertex-ai/docs/pipelines/schedule-pipeline-run).
Announcement Vertex AI Tensorboard pricing has changed from a per-user monthly license of $300 per month to $10 GiB per month for storage of your logs. This means no more subscription fees. You only pay for the storage you've used. See the [Vertex AI Tensorboard: Delete Outdated Tensorboard Experiments](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/experiments/delete_outdated_tensorboard_experiments.ipynb) tutorial for how to manage storage.

## July 28, 2023

**Vertex AI** v1 Change The `learning_rate` parameter in generative AI model tuning is now `learning_rate_multiplier`.
To use the model's or tuning method's default learning rate, use the default
`learning_rate_multiplier` value of `1.0`.

If you haven't configured `learning_rate` before, no action is needed.
If using `tuning_method=tune_v2` with the v2.0.0 pipeline template
(Python SDK v1.28.1+), the recommended learning rate is 0.0002. To convert your
custom `learning_rate` to `learning_rate_multiplier`, calculate as follows:

    learning_rate_multiplier = custom_learning_rate_value / 0.0002

## July 19, 2023

**Agent Platform Workbench** v2 Feature Vertex AI Workbench instances are now available in [Preview](https://cloud.google.com/products/#product-launch-stages). Vertex AI Workbench instances combine features from managed notebooks and user-managed notebooks to provide a robust data science solution. Supported features include:

- Idle timeout
- BigQuery and Cloud Storage integrations
- End-user and service account authentication
- VPC Service Controls
- Customer managed encryption keys (CMEK)
- Health status monitoring
- Run notebooks on a schedule
- Dataproc integration

To get started, see [Introduction to Vertex AI Workbench instances](https://docs.cloud.google.com/vertex-ai/docs/workbench/instances/introduction).

## July 18, 2023

**Vertex AI** v1 Feature Model tuning updates for text-bison:

- Upgraded tuning pipeline now offers more efficient tuning and better performance on text-bison.
- New `learning_rate` parameter lets you adjust the step size at each iteration.

For details, see
[Tune language foundation models](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/models/tune-models).

## July 17, 2023

**Vertex AI** v1 Feature Imagen on Vertex AI now offers the following [Generally Available (GA)](https://cloud.google.com/products#product-launch-stages) features:

- [Image generation (text-to-image generation)](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/image/generate-images)\*
- [Image editing](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/image/edit-images)\*
- [Image visual captioning](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/image/image-captioning)
- [Visual Question Answering (VQA)](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/image/visual-question-answering)

\* Restricted access feature.

For more information about Imagen or how to get access to restricted GA or Preview features, see the
[Imagen on Vertex AI overview](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/image/overview).
Feature The Vertex AI PaLM API has added support for the following languages:

- Spanish (es)
- Korean (ko)
- Hindi (hi)
- Chinese (zh)

For the complete list of supported languages, see
[Supported languages](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/learn/models#language_support).
Feature Imagen now supports human face generation for the following features:

- [Image generation (text-to-image generation)](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/image/generate-images)\*
- [Image editing](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/image/edit-images)\*

\* Restricted access feature.

Human face generation is enabled by default, except for images with children
and/or celebrities. For more information, see the
[usage guidelines](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/image/usage-guidelines).

## July 13, 2023

**Vertex AI** v1 Feature Support for [batch text (`text-bison`) requests](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/text/batch-prediction-genai)
is now available in ([GA](https://cloud.google.com/products/#product-launch-stages)).
You can review pricing for the `chat-bison` model at
[Vertex AI pricing page](https://cloud.google.com/vertex-ai/pricing#generative_ai_models).

## July 10, 2023

**Vertex AI** v1 Feature Support for PaLM 2 for [Chat (`chat-bison`)](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/chat/chat-prompts)
is now available in ([GA](https://cloud.google.com/products/#product-launch-stages)).
You can review pricing for the `chat-bison` model at
[Vertex AI pricing page](https://cloud.google.com/vertex-ai/pricing#generative_ai_models).

## July 07, 2023

**Vertex AI** v1 Announcement [GCSFuse support for custom training](https://docs.cloud.google.com/vertex-ai/docs/training/cloud-storage-file-system) is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

## July 06, 2023

**Vertex AI** v1 Feature **Vertex AI model evaluation** is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)) with the following new Preview features:

- Model evaluation with sliced metrics.
- Model evaluation with fairness and bias metrics.
- Vision error analysis for AutoML image classification models.

## June 30, 2023

**Vertex AI** v1 Feature **Vertex Explainable AI**

Support for [example-based explanations](https://docs.cloud.google.com/vertex-ai/docs/explainable-ai/overview.md.txt#example-based) is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)).
Deprecated [Vertex AI data labeling](https://docs.cloud.google.com/vertex-ai/docs/datasets/data-labeling-job) is deprecated and will no longer be available on Google Cloud after July 1, 2024. For new labeling tasks, you can use [add labels using the Google Cloud console](https://cloud.google.com/vertex-ai/docs/datasets/label-using-console) or access data labeling solutions from our partners in the [Google Cloud console marketplace](https://console.cloud.google.com/marketplace/), such as Labelbox and Snorkel.

## June 29, 2023

**Vertex AI** v1 Feature **Vertex AI Codey APIs**

The [Vertex AI Codey APIs](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/code/code-models-overview) are now generally available ([GA](https://cloud.google.com/products#product-launch-stages)).
Use the Codey APIs to create solutions with code generation, code completion, and code chat. Because the Vertex AI Codey APIs are GA, you incur usage costs if you use them. To learn about pricing, see the [Generative AI support on Vertex AI pricing page](https://cloud.google.com/vertex-ai/pricing#generative_ai_models).

The models in this release include:

- `code-bison` (code generation)
- `codechat-bison` (multi-turn code chat)
- `code-gecko` (code completion)

The maximum tokens for input was increased from 4,096 to 6,144 tokens for `code-bison` and `codechat-bison` to allow longer prompts and chat history. The maximum tokens for output was increased from 1,024 to 2,048 for `code-bison` and `codechat-bison` to allow for longer responses.

Additional programming languages are supported. For more information, see [Supported coding languages](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/code/code-models-overview#supported_coding_languages).

Several fine-tuning datasets were removed from the `code-bison` and `codechat-bison` models to implement the following improvements:

- Excessive chattiness.
- Artifacting, such as NBSP (non-breaking space) characters.
- Low quality code responses.

To learn about cloud horizontals, please see [Vertex AI certifications](https://docs.cloud.google.com/vertex-ai/docs/general/vertexai-compliance).
Feature Vertex AI Pipeline task-level logs are now generally available ([GA](https://cloud.google.com/products#product-launch-stages)) in Cloud Logging. Additionally, from Cloud Logging you can route pipeline logs to a Pub/Sub sink to power your event-driven architecture. For more information, see [View pipeline job logs](https://docs.cloud.google.com/vertex-ai/docs/pipelines/logging).

## June 26, 2023

**Agent Platform Workbench** v1 Feature **M109 release**

The M109 release of Vertex AI Workbench user-managed notebooks includes the following:

- PyTorch 2.0 with Python 3.10 and CUDA 11.8 user-managed notebooks instances are now available.
- Miscellaneous software updates.
Feature The M109 release of Vertex AI Workbench managed notebooks includes the following:

- Fixed a bug that caused high cpu utilization due to excessive internal diagnostic tool processes.
- Fixed a bug that was showing incorrect kernel image icons in the Jupyterlab launcher.

## June 20, 2023

**Vertex AI** v1 Feature A100 80GB accelerators are now generally available ([GA](https://cloud.google.com/products#product-launch-stages)) for custom training jobs in the following regions:

- asia-southeast1
- europe-west4
- us-central1
- us-east4

For more information, see [Locations](https://docs.cloud.google.com/vertex-ai/docs/general/locations#accelerators).
Announcement The Kubeflow Pipelines (KFP) SDK v2 is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)). KFP SDK v2 introduces several improvements for authoring pipelines and is fully supported by Vertex AI Pipelines.

To learn more about the changes in KFP v2, see the [KFP v2 Release Notes](https://github.com/kubeflow/pipelines/blob/master/sdk/RELEASE.md#200) and [KFP v2 migration guide](https://www.kubeflow.org/docs/components/pipelines/v2/migration/).
Announcement The Google Cloud Pipeline Components (GCPC) SDK v2 is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)). GCPC v2 introduces support for the KFP v2 SDK and is fully supported by Vertex AI Pipelines.

To learn more about the updates in the latest version of the GCPC SDK, see the [Google Cloud Pipelines Components Release Notes](https://google-cloud-pipeline-components.readthedocs.io/en/google-cloud-pipeline-components-2.0.0/release.html).

## June 15, 2023

**Vertex AI** v1 Feature The `chat-bison@001` model has been updated to better follow instructions in the `context` field. For details, on how to create chat prompts for `chat-bison@001`, see [Design chat prompts](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/chat/chat-prompts).

## June 09, 2023

**Vertex AI** v1 Announcement **HIPAA compliance for Generative AI on Vertex AI**

Generative AI support on Vertex AI now supports HIPAA compliance. The coverage includes components of the [Model Garden](https://docs.cloud.google.com/vertex-ai/docs/start/explore-models) and [Generative AI Studio](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/learn/generative-ai-studio).

To learn more about Vertex certifications, see [Vertex AI features](https://docs.cloud.google.com/vertex-ai/docs/general/features) and [Vertex AI certifications](https://docs.cloud.google.com/vertex-ai/docs/general/vertexai-compliance).

## June 07, 2023

**Vertex AI** v1 Feature **Vertex AI Codey APIs**

The Vertex AI Codey APIs are now in [Preview](https://cloud.google.com/products#product-launch-stages).
With the Codey API, code generation, code completion, and code chat APIs can be used from any Google Cloud project without allowlisting. The APIs can be accessed from the
`us-central1` region. The Codey APIs can be used in the Generative AI studio or
programmatically in REST commands.

To get started, see the [Code models overview](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/code/code-models-overview).
Feature **Vertex AI Model Garden**

The Vertex AI Model Garden is [generally available (GA)](https://cloud.google.com/products#product-launch-stages)). Model Garden is a platform that helps you to discover, test, customize, and deploy Vertex AI and select OSS models. These models range from tunable to task-specific. You can find these models on the Model Garden page in the Google Cloud console.

To get started, see [Explore AI models and APIs in Model Garden](https://docs.cloud.google.com/vertex-ai/docs/start/explore-models#create-a-vision-app).
Feature **PaLM Text and Embeddings APIs, and Generative AI Studio**

The Generative AI support on Vertex AI is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)).
With this feature launch, you can leverage the PaLM API to generate
AI models that you can test, tune, and deploy in your AI-powered applications.
With the GA of these features, you will incur usage costs if you use the
`text-bison` and `textembedding-gecko` PaLM APIs. To learn about pricing, see
the [Vertex AI pricing page](https://cloud.google.com/vertex-ai/pricing#generative_ai_models).

Features and models in this release include:

- PaLM 2 for Text: `text-bison`
- Embedding for Text: `textembedding-gecko`
- Generative AI Studio for Language

> [!IMPORTANT]
> **Important:** With this GA launch, standard security and compliance for Vertex AI is not yet available to Generative AI. To learn more, see [Vertex AI features](https://docs.cloud.google.com/vertex-ai/docs/general/features) and [Vertex AI certifications](https://docs.cloud.google.com/vertex-ai/docs/general/vertexai-compliance).

## June 01, 2023

**Vertex AI** v1 Feature **Vertex Prediction**

You can now specify a multi-region BigQuery table as the input or output to a [batch prediction request](https://docs.cloud.google.com/vertex-ai/docs/predictions/get-predictions#get_batch_predictions).

## May 18, 2023

**Vertex AI** v1 Feature **Vertex Prediction**

You can co-host models on the same VM from the Google Cloud console. Previously, this capability was available only from the REST API. For more information, see [Share resources across deployments](https://docs.cloud.google.com/vertex-ai/docs/predictions/model-co-hosting).

## May 16, 2023

**Vertex AI** v1 Feature Vertex AI custom training now supports deep integration with Vertex AI Experiments. You can submit training jobs with autologging enabled to automatically log parameters and model performance metrics. For more information, see [Run training job with experiment tracking](https://docs.cloud.google.com/vertex-ai/docs/experiments/run-training-job-experiments)
Feature The scheduler API for Vertex AI Pipelines is now available in [Preview](https://cloud.google.com/products#product-launch-stages). You can schedule recurring pipeline runs in Vertex AI by specifying a frequency, start time (optional), and end time (optional). For more information, see [Schedule a pipeline run with scheduler API](https://docs.cloud.google.com/vertex-ai/docs/pipelines/schedule-pipeline-run).

## May 10, 2023

**Vertex AI** v1 Feature **Generative AI Support for Vertex AI**

Generative AI Support for Vertex AI is now available in [Preview](https://cloud.google.com/products#product-launch-stages). With this feature launch, you can leverage the Vertex AI PaLM API to generate AI models that you can test, tune, and deploy in your AI-powered applications.

Features and models in this release include:

- PaLM 2 for Text: text-bison@001
- PaLM 2 for Chat: chat-bison@001
- Embedding for Text: textembedding-gecko@001
- Generative AI Studio for Language
- Tuning for PaLM 2
- Vertex AI SDK v1.25, which includes new features such as TextGenerationModel(text-bison@001), ChatModel(chat-bison@001), TextEmbeddingModel(textembedding-gecko@001)

You can interact with the generative AI features on Vertex AI by using Vertex AI Studio in the Google Cloud console, the Vertex AI API, and the Vertex AI SDK for Python.

- Learn more about [Generative AI Support for Vertex AI](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/learn/overview)
- See an [Introduction to Generative AI Studio](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/learn/generative-ai-studio)
- Get started with a [Generative AI Studio quickstart](https://docs.cloud.google.com/vertex-ai/docs/generative-ai/start/quickstarts/quickstart)
Feature **Vertex AI Model Garden**

The Vertex AI Model Garden is available in [Preview](https://cloud.google.com/products#product-launch-stages). The Model Garden is a platform that helps you to discover, test, customize, and deploy Vertex AI and select OSS models. These models range from tunable to task-specific. You can find these models on the Model Garden page in the Google Cloud console.

- To get started, see [Explore AI models and APIs in Model Garden](https://docs.cloud.google.com/vertex-ai/docs/start/explore-models#create-a-vision-app).

## May 09, 2023

**Vertex AI** v1 Feature **Vertex AI Prediction**

You can now use [G2 accelerator-optimized machine types](https://docs.cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) to serve predictions. Each G2 machine has a fixed number of NVIDIA L4 GPUs attached.

## May 04, 2023

**Agent Platform Workbench** v1 Feature **M108 release**

The M108 release of Vertex AI Workbench user-managed notebooks includes the following:

- Miscellaneous software updates.

## April 14, 2023

**Vertex AI** v1 Feature **Vertex AI Prediction**

You can now update some scaling and container logging configuration settings on a [`DeployedModel`](https://docs.cloud.google.com/vertex-ai/docs/reference/rest/v1/projects.locations.endpoints#deployedmodel) without undeploying and redeploying it to an endpoint.

For more information, see [update the scaling configuration](https://docs.cloud.google.com/vertex-ai/docs/predictions/configure-compute#update_the_scaling_configuration) and [container logging](https://docs.cloud.google.com/vertex-ai/docs/predictions/online-prediction-logging).

## April 13, 2023

**Agent Platform Workbench** v1 Feature **M107 release**

The M107 release of Vertex AI Workbench user-managed notebooks includes the following:

- Fixed a bug that displayed the wrong version of the JupyterLab user interface.
- Fixed a bug where a cron job for the diagnostic tool was added at every restart.
- Miscellaneous software updates.
**Vertex AI** v1 Announcement The Timeseries Insights API is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)). With the Timeseries Insights API, you can forecast and detect anomalies over billions of events in real time. For more information, see [Timeseries Insights](https://docs.cloud.google.com/timeseries-insights/docs/overview).

## April 06, 2023

**Agent Platform Workbench** v1 Feature **M106 release**

The M106 release of Vertex AI Workbench user-managed notebooks includes the following:

- Rolled back a previous change in which Jupyter dependencies were located in a separate Conda environment.
- Fixed a bug in which kernels used by notebooks did not contain the specified machine learning frameworks.
- Miscellaneous software updates.

## April 04, 2023

**Vertex AI** v1 Feature Vertex AI Pipelines is now integrated with [Cloud Asset Inventory](https://docs.cloud.google.com/asset-inventory/docs/overview) service. You can use Cloud Asset Inventory to search, export, monitor, and analyze pipeline resources and metadata, and also view the resource history.
Feature **Vertex AI Prediction**

You can now view logs for Vertex AI [Batch Prediction](https://docs.cloud.google.com/vertex-ai/docs/predictions/overview#batch_predictions) jobs in [Cloud Logging](https://docs.cloud.google.com/logging/docs/overview.md.txt).
Feature The Vertex AI Matching Engine service now offers [Preview](https://cloud.google.com/products#product-launch-stages) support for deploying an index to a public endpoint. For information about how to get started, see [Matching Engine Setup](https://docs.cloud.google.com/vertex-ai/docs/matching-engine/match-eng-setup/match-eng-setup).

## April 03, 2023

**Vertex AI** v1 Feature The Vertex AI Model Registry now offers [Preview](https://cloud.google.com/products#product-launch-stages) support for model copy between regions. For information about how to copy your model between regions, see [Copy models in Model Registry](https://docs.cloud.google.com/vertex-ai/docs/model-registry/copy-model).

## March 31, 2023

**Agent Platform Workbench** v1 Feature **M105 release**

The M105 release of Vertex AI Workbench user-managed notebooks includes the following:

- The following user-managed notebooks images are now available with Python 3.10 on Debian 11:

  - TensorFlow 2.11 CPU (`tf-2-11-cpu-debian-11-py310`)
  - TensorFlow 2.11 GPU with Cuda 11.3 (`tf-2-11-cu113-notebooks-debian-11-py310`)
  - PyTorch 1.13 with Cuda 11.3 (`pytorch-1-13-cu113-notebooks-debian-11-py310`)
  - Base CPU (`common-cpu-notebooks-debian-11-py310`)
  - Base GPU with Cuda 11.3 (`common-cu113-notebooks-debian11-py310`)
- The following user-managed notebooks images are now available with Python 3.9 on Debian 11:

  - TensorFlow 2.6 CPU (`tf-2-6-cpu-notebooks-debian-11-py39`)
  - TensorFlow 2.6 GPU with Cuda 11.3 (`tf-2-6-cu113-notebooks-debian-11-py39`)
- Jupyter-related libraries have been moved to a different Conda environment, separate from the one containing machine learning frameworks and base software libraries.

## March 28, 2023

**Vertex AI** v1 Feature Vertex AI Pipelines cost showback with billing labels is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)). You can now use billing labels to review the cost of a pipeline run, along with the cost of individual resources generated from Google Cloud Pipeline Components in the pipeline run. For more information, see [Understand pipeline run costs](https://docs.cloud.google.com/vertex-ai/docs/pipelines/understand-pipeline-cost-labels).

## March 27, 2023

**Agent Platform Workbench** v1 Feature **M105 release**

The M105 release of Vertex AI Workbench managed notebooks includes the following:

- Fixed an issue wherein a runtime with idle shutdown enabled doesn't detect activity and shuts down.
- Fixed an issue wherein the runtime data disk runs out of space and prevents access.
- Fixed an issue wherein end user credentials are not preserved after shutdown.
- Changed Health Agent logging levels from `DEBUG` to `INFO`.

## March 21, 2023

**Vertex AI** v1 Feature Vertex AI supports running Explainable AI on certain types of BQML models when they are added to the Vertex AI Model Registry [(GA)](https://cloud.google.com/products#product-launch-stages). To learn more, see [Explainable AI for BigQuery ML models](https://docs.cloud.google.com/bigquery-ml/docs/vertex-xai).
Feature **Vertex AI Feature Store**

The ability to delete feature values from an entity type is now generally available ([GA](https://cloud.google.com/products/#product-launch-stages)). The following features are available:

- [Delete feature values from specified entities](https://docs.cloud.google.com/vertex-ai/docs/featurestore/delete-feature-values.md.txt#delete_from_entities)
- [Delete feature values from specified features within a time range](https://docs.cloud.google.com/vertex-ai/docs/featurestore/delete-feature-values.md.txt#delete_from_features_timerange)

Links to additional resources:

- [Learn more about the `delete_feature_values` method in the Vertex AI SDK for Python.](https://docs.cloud.google.com/python/docs/reference/aiplatform/latest/google.cloud.aiplatform_v1.services.featurestore_service.FeaturestoreServiceClient.md.txt#google_cloud_aiplatform_v1_services_featurestore_service_FeaturestoreServiceClient_delete_feature_values)
- [View the Python code sample on GitHub](https://github.com/googleapis/python-aiplatform/blob/main/samples/generated_samples/aiplatform_v1_generated_featurestore_service_delete_feature_values_sync.py)
- [View the Python (Async) code sample on GitHub](https://github.com/googleapis/python-aiplatform/blob/main/samples/generated_samples/aiplatform_v1_generated_featurestore_service_delete_feature_values_async.py)

## March 20, 2023

**Vertex AI** v1 Feature **Vertex AI Prediction**

You can now use [N2, N2D, C2, and C2D machine types](https://docs.cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) to serve predictions.

## March 16, 2023

**Agent Platform Workbench** v1 Feature **M104 release**

The M104 release of Vertex AI Workbench user-managed notebooks includes the following:

- Fixed a regression in which `jupyter-user` metadata was ignored.
- Enabled access to the Jupyter Gateway Client configuration by using the `notebook-enable-gateway-client` and `gateway-client-url` metadata tags.
- Added the following packages:
  - google-cloud-artifact-registry
  - google-cloud-bigquery-storage
  - google-cloud-language
  - keyring
  - keyrings.google-artifactregistry-auth
- Fixed a bug in which curl could not find the right SSL certificate path by default.
Change TensorFlow Enterprise 2.1 has reached the end of its support period. See [Version details](https://docs.cloud.google.com/tensorflow-enterprise/docs/overview.md.txt#version_details).

## March 03, 2023

**Vertex AI** v1 Change Pre-built containers to perform custom training with [TensorFlow 2.11](https://cloud.google.com/ai-platform-unified/docs/training/pre-built-containers#tensorflow), [PyTorch 1.12](https://cloud.google.com/vertex-ai/docs/training/pre-built-containers#pytorch), or [PyTorch 1.13](https://cloud.google.com/vertex-ai/docs/training/pre-built-containers#pytorch) are now generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

## February 28, 2023

**Vertex AI** v1 Change A new [custom training overview](https://docs.cloud.google.com/vertex-ai/docs/training/overview) page is available. The new overview page covers the following topics:

- What is custom training?
- Benefits of custom training on Vertex AI.
- How custom training works.
- Custom training workflow.

## February 21, 2023

**Agent Platform Workbench** v1 Feature **M104 update**

This update of the M104 release of Vertex AI Workbench managed notebooks includes the following:

- Fixed a bug where local and remote kernels are not displayed. This happens when remote kernels are not accessible.
- Minor bug fixes and improvements.

## February 14, 2023

**Vertex AI** v1 Feature **Vertex AI Prediction**

[Pre-built PyTorch containers](https://docs.cloud.google.com/vertex-ai/docs/predictions/pre-built-containers#pytorch) for serving predictions from PyTorch models is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).
Feature Vertex AI Matching Engine now supports Private Service Connect in [Preview](https://cloud.google.com/products#product-launch-stages). To learn how to set up a
a Private Service Connect instance, see [Using Private Service Connect](https://docs.cloud.google.com/vertex-ai/docs/matching-engine/match-eng-setup/private-service-connect).

## February 13, 2023

**Vertex AI** v1 Feature Support for resource-level IAM policies for Vertex AI `featurestore` and `entityType` resources is generally available ([GA](https://cloud.google.com/products#product-launch-stages)). For more information, see [Control access to resources](https://docs.cloud.google.com/vertex-ai/docs/featurestore/resource-policy.md.txt).

## February 10, 2023

**Vertex AI** v1 Fixed When performing distributed training, Vertex AI properly sets the primary replica in `CLUSTER_SPEC` as `workerpool0` instead of `chief`. For details, see [Format CLUSTER_SPEC](https://docs.cloud.google.com/vertex-ai/docs/training/distributed-training#cluster-spec-format).

## February 09, 2023

**Agent Platform Workbench** v1 Feature **M104 release**

The M104 release of Vertex AI Workbench managed notebooks includes the following:

- Added a fix for a security vulnerability in single-user managed notebooks instances.
- Made enhancements to the network selection user experience in the managed notebooks executor.
- Minor bug fixes and improvements.

## February 06, 2023

**Vertex AI** v1 Feature The Vertex AI Pipelines **Template Gallery** is now available in [Preview](https://cloud.google.com/products#product-launch-stages). You can bootstrap your MLOps workflows with Google-authored pipeline and component templates. For more information, see [Use a prebuilt template from the Template Gallery](https://docs.cloud.google.com/vertex-ai/docs/pipelines/use-template-gallery).

## January 30, 2023

**Agent Platform Workbench** v1 Feature **M103 release**

The M103 release of Vertex AI Workbench user-managed notebooks includes the following:

- Fixed a bug in which a warning tells the user to run `jupyter lab build` when creating a new instance.
- Upgraded PyTorch to 1.13.1.
- Minor bug fixes and improvements.

## January 26, 2023

**Vertex AI** v1 Feature **Tabular Workflow for End-to-End AutoML** is generally available ([GA](https://cloud.google.com/products#product-launch-stages)). For documentation, refer to [Tabular Workflow for End-to-End AutoML](https://docs.cloud.google.com/vertex-ai/docs/tabular-data/tabular-workflows/e2e-automl).

## January 18, 2023

**Vertex AI** v1 Change **Vertex AI Explainability**

When uploading TensorFlow 2 models, the [`ExplanationMetadata`](https://docs.cloud.google.com/vertex-ai/docs/reference/rest/v1/ExplanationSpec#ExplanationMetadata) field is now optional, making it easier to configure your model for explainability. For more information, see [Import a model with an `explanationSpec` field](https://docs.cloud.google.com/vertex-ai/docs/explainable-ai/configuring-explanations-feature-based.md.txt#import-model-example).

## January 11, 2023

**Vertex AI** v1 Change Vertex AI Matching Engine is available in the following regions:

- `us-west2` -- (Los Angeles)
- `us-west3` -- (Salt Lake City)
- `northamerica-northeast1` -- (Montréal)
- `northamerica-northeast2` -- (Toronto)
- `europe-central2` -- (Warsaw)
- `europe-west2` -- (London)
- `europe-west3` -- (Frankfurt)
- `europe-west6` -- (Zurich)
- `asia-east1` -- (Taiwan)
- `Asia-east2` -- (Hong Kong)
- `me-west1` -- (Tel aviv)

To see all of the available locations for Matching Engine, see the [Vertex AI Locations page](https://docs.cloud.google.com/vertex-ai/docs/general/locations).

## December 20, 2022

**Vertex AI** v1 Feature **Vertex AI TensorFlow Profiler**

Vertex AI TensorFlow Profiler is generally available [GA](https://cloud.google.com/products#section-23). You can use TensorFlow Profiler to debug model training performance for your custom training jobs.

For details, see [Profile model training performance using Profiler](https://docs.cloud.google.com/vertex-ai/docs/experiments/tensorboard-profiler).
Feature **Vertex AI Matching Engine**

[Vertex AI Matching Engine](https://docs.cloud.google.com/vertex-ai/docs/matching-engine/overview) now offers [General Availability](https://cloud.google.com/products/#product-launch-stages) support for updating your indices using Streaming Update, which is real-time indexing for the Approximate Nearest Neighbor (ANN) service.
Feature Vertex AI Feature Store [streaming ingestion](https://docs.cloud.google.com/vertex-ai/docs/featurestore/ingesting-stream.md.txt) is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)).
Change You can now override the default data retention limit of 4000 days for the online store and the offline store in Vertex AI Feature Store.

- You can set the data retention limit for the online store at the featurestore level.
- You can set the data retention limit for the offline store at the entity type level.

## December 15, 2022

**Agent Platform Workbench** v1 Feature **M102 release**

The M102 release of Vertex AI Workbench user-managed notebooks includes the following:

- TensorFlow 2.11 is now available.
- PyTorch 1.13 is now available.
- Regular security patches and package upgrades.

## December 09, 2022

**Agent Platform Workbench** v1 Feature **M101 release**

The M101 release of Vertex AI Workbench includes the following:

- TensorFlow patch version upgrades:
  - From 2.8.3 to 2.8.4.
  - From 2.9.2 to 2.9.3.
  - From 2.10.0 to 2.10.1.
- TensorFlow 1.15 on Vertex AI Workbench is now deprecated.
- Added `*.notebooks.cloud.google.com` as part of the domains required for users to access Notebooks API. Removed `*.datalab.cloud.google.com`.
- Regular security patches and package upgrades.

## December 05, 2022

**Vertex AI** v1 Feature The Pipeline Templates feature is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)). The **Your Templates** tab is supported by Artifact Registry and allows you to publish and curate pipeline and component templatess. For documentation, refer to [Create, upload, and use a pipeline template](https://docs.cloud.google.com/vertex-ai/docs/pipelines/create-pipeline-template).

## November 30, 2022

**Vertex AI** v1 Feature **AutoML image model updates**

AutoML image classification and object detection now support a higher-accuracy model type. This model is available in [Preview](https://cloud.google.com/products#product-launch-stages).

For information about how to train a model using the higher accuracy model type, see [Begin AutoML model training](https://docs.cloud.google.com/vertex-ai/docs/tutorials/image-recognition-automl/training#2_begin_automl_model_training).

Batch prediction is currently not supported for this model type.
Feature Cloud Logging for Vertex AI Pipelines is now generally available ([GA](https://cloud.google.com/products/#product-launch-stages)). For more information, see [View pipeline job logs](https://docs.cloud.google.com/vertex-ai/docs/pipelines/logging).

## November 18, 2022

**Vertex AI** v1 Feature **Vertex AI Prediction**

You can now perform some simple filtering and transformation on the batch input in your [`BatchPredictionJob`](https://docs.cloud.google.com/vertex-ai/docs/reference/rest/v1beta1/projects.locations.batchPredictionJobs) requests without having to write any code in the prediction container. This feature is in [Preview](https://cloud.google.com/products/#product-launch-stages). For more information, see [Filter and transform input data](https://docs.cloud.google.com/vertex-ai/docs/predictions/get-predictions#filter_and_transform_input_data_preview).

## November 17, 2022

**Vertex AI** v1 Feature The Vertex AI Pipelines email notification component is now generally available ([GA](https://cloud.google.com/products/#product-launch-stages)). This component enables you to configure your pipeline to send up to three emails upon success or failure of a pipeline run. For more information, see [Configure email notifications](https://docs.cloud.google.com/vertex-ai/docs/pipelines/email-notifications) and the [Email notification component](https://docs.cloud.google.com/vertex-ai/docs/pipelines/email-notification-component).

## November 16, 2022

**Vertex AI** v1 Feature Vertex AI has added support for the following regions:

- `us-west3` (Salt Lake City)
- `europe-central2` (Warsaw)
- `asia-southeast2` (Jakarta)
- `me-west1` (Tel aviv)

Some features of Vertex AI are not supported in these regions. Check feature availability for all regions on the [Vertex AI Locations page](https://docs.cloud.google.com/vertex-ai/docs/general/locations).

## November 10, 2022

**Vertex AI** v1 Feature **AutoML Image Classification Error Analysis**

Error analysis allows you to examine error cases after training a model from within the model evaluation page. This feature is available in [Preview](https://cloud.google.com/products#product-launch-stages).

For each image you can inspect similar images from the training set to help identify the following:

- Label inconsistencies between visually similar images
- Outliers if a test sample has no visually similar images in the training set

After fixing any data issues, you can retrain the model to improve model performance.

## November 09, 2022

**Vertex AI** v1 Feature Feature Transform Engine is available in [Preview](https://cloud.google.com/products#product-launch-stages). For documentation, refer to [Feature engineering](https://docs.cloud.google.com/vertex-ai/docs/tabular-data/tabular-workflows/feature-engineering).

## November 08, 2022

**Agent Platform Workbench** v1 Feature **M100 release**

The M100 release of Vertex AI Workbench includes the following:

- Fixed a bug that prevented an instance with a GPU from starting.
- Regular package updates.
- Miscellaneous bug and display fixes.
Fixed Fixed a server-side request forgery (SSRF) vulnerability. Previous versions of managed notebooks and user-managed notebooks instances still contain the vulnerability. It is recommended that you [migrate your data](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/migrate) to a new instance.

## November 04, 2022

**Vertex AI** v1 Feature **Vertex ML Metadata**

You can now [filter contexts, executions, and artifacts by association and attribution](https://docs.cloud.google.com/vertex-ai/docs/ml-metadata/analyzing#filter_contexts_executions_and_artifacts_by_association_and_attribution).
Feature **Vertex AI Prediction**

You can now use [A2 machine types](https://docs.cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) to serve predictions.
Feature Custom training on Vertex AI now supports [NVIDIA A100 80GB GPUs](https://docs.cloud.google.com/compute/docs/gpus.md.txt#a100-80gb) on `a2-ultragpu-1g/2g/4g/8g` machines. For details, see [Configure compute resources for custom training](https://docs.cloud.google.com/vertex-ai/docs/training/configure-compute).

## November 03, 2022

**Vertex AI** v1 Feature **Vertex AI Prediction**

[Custom prediction routines (CPR)](https://docs.cloud.google.com/vertex-ai/docs/predictions/custom-prediction-routines) are now [Generally Available](https://cloud.google.com/products/#product-launch-stages). CPR lets you easily build custom containers for prediction with pre/post processing support.

## October 27, 2022

**Vertex AI** v1 Feature **Vertex AI Prediction**

You can now use [E2 machine types](https://docs.cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) to serve predictions.

## October 25, 2022

**Agent Platform Workbench** v1beta1 Breaking The `v1beta1` version of the Notebooks API is scheduled for removal no earlier than January 16, 2023. After this date, you must use Notebooks API `v1` to manage Vertex AI Workbench resources.

## October 18, 2022

**Agent Platform Workbench** v1 Feature **M98 release**

The M98 release of Vertex AI Workbench managed notebooks includes the following:

- Upgraded Go from 1.16.5 to 1.19.2.
- Upgraded R from 4.1 to 4.2.
- Upgraded JupyterLab from 3.2 to 3.4.
- Miscellaneous bug and display fixes.
- Added a fix for the BigQuery SQL editor to run queries correctly in non-US locations.
- Regular package updates.

[Learn more about managed notebooks versions](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/versions).

## October 12, 2022

**Vertex AI** v1 Feature Tabular Workflow for TabNet Training is available in [Preview](https://cloud.google.com/products#product-launch-stages). For documentation, refer to [Tabular Workflows for TabNet Training](https://docs.cloud.google.com/vertex-ai/docs/tabular-data/tabular-workflows/tabnet.md.txt).

Tabular Workflow for Wide \& Deep Training is available in [Preview](https://cloud.google.com/products#product-launch-stages). For documentation, refer to [Tabular Workflow for Wide \& Deep Training](https://docs.cloud.google.com/vertex-ai/docs/tabular-data/tabular-workflows/wide-and-deep).

## October 11, 2022

**Vertex AI** v1 Feature Vertex AI Feature Store [streaming ingestion](https://docs.cloud.google.com/vertex-ai/docs/featurestore/ingesting-stream.md.txt) is available in [Preview](https://cloud.google.com/products#product-launch-stages).

## October 10, 2022

**Vertex AI** v1 Feature The **Vertex AI Model Registry and BigQuery ML** integration is generally available ([GA](https://cloud.google.com/products#product-launch-stages)). With this integration, [BigQuery ML](https://docs.cloud.google.com/bigquery-ml/docs/managing-models-vertex) models can be managed alongside other ML models in Vertex AI to easily version, evaluate, and deploy for prediction.
Feature **The Vertex AI Model Registry** is generally available ([GA](https://cloud.google.com/products#product-launch-stages)). [Vertex AI Model Registry](http://cloud.google.com/vertex-ai/docs/model-registry/introduction) is a searchable repository where you can manage the lifecycle of your ML models. From the Vertex AI Model Registry, you can better organize your models, train new versions, and deploy directly to endpoints.

## October 06, 2022

**Vertex AI** v1 Feature **Incrementally train an AutoML model**

You can now incrementally train an AutoML image classification or object detection model by selecting a previously trained model. This feature is in [Preview](https://cloud.google.com%20/products/#product-launch-stages). For more information, see [Train an AutoML image classification model](https://docs.cloud.google.com/vertex-ai/docs/tutorials/image-recognition-automl/training).

## October 05, 2022

**Vertex AI** v1 Feature **Vertex AI Feature Store**

The ability to [delete feature values](https://docs.cloud.google.com/vertex-ai/docs/featurestore/delete-feature-values.md.txt) from an entity type is now available in [Preview](https://cloud.google.com/products#product-launch-stages). The following features are available:

- [Delete feature values from specified entities](https://docs.cloud.google.com/vertex-ai/docs/featurestore/delete-feature-values.md.txt#delete_from_entities)
- [Delete feature values from specified features within a time range](https://docs.cloud.google.com/vertex-ai/docs/featurestore/delete-feature-values.md.txt#delete_from_features_timerange)

Links to additional resources:

- [Learn more about the `delete_feature_values` method in the Vertex AI SDK for Python.](https://docs.cloud.google.com/python/docs/reference/aiplatform/latest/google.cloud.aiplatform_v1beta1.services.featurestore_service.FeaturestoreServiceClient.md.txt#google_cloud_aiplatform_v1beta1_services_featurestore_service_FeaturestoreServiceClient_delete_feature_values)
- [View the Python code sample on GitHub](https://github.com/googleapis/python-aiplatform/blob/9a506eeae6be3fa33b262dc7782d0bb4f10bf765/samples/generated_samples/aiplatform_v1beta1_generated_featurestore_service_delete_feature_values_sync.py)
- [View the Python (Async) code sample on GitHub](https://github.com/googleapis/python-aiplatform/blob/9a506eeae6be3fa33b262dc7782d0bb4f10bf765/samples/generated_samples/aiplatform_v1beta1_generated_featurestore_service_delete_feature_values_async.py)

## October 04, 2022

**Vertex AI** v1 Feature **Vertex AI model evaluation** is now available in Preview. [Model evaluation](https://docs.cloud.google.com/vertex-ai/docs/evaluation/introduction) provides model evaluation metrics, such as precision and recall, to help you
determine the performance of your models.

## September 26, 2022

**Vertex AI** v1 Feature **Vertex AI Model Monitoring**

[Vertex AI Model Monitoring](https://docs.cloud.google.com/vertex-ai/docs/model-monitoring) now offers [Preview](https://cloud.google.com/products#product-launch-stages) support for batch prediction jobs. For more details, see [Vertex AI Model Monitoring for batch predictions](https://docs.cloud.google.com/vertex-ai/docs/model-monitoring/model-monitoring-batch-predictions).
Feature **Vertex AI Feature Store**

[Feature value monitoring](https://docs.cloud.google.com/vertex-ai/docs/featurestore/monitoring.md.txt#feature) is now generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

## September 22, 2022

**Vertex AI** v1 Feature **Vertex AI Matching Engine**

[Vertex AI Matching Engine](https://docs.cloud.google.com/vertex-ai/docs/matching-engine/overview) now offers [Preview](https://cloud.google.com/products/#product-launch-stages) support for updating your indices using Streaming Update, which is real-time indexing for the Approximate Nearest Neighbor (ANN) service.

## September 20, 2022

**Agent Platform Workbench** v1 Feature **M96 release**

The M96 release of Vertex AI Workbench managed notebooks includes the following:

- Fixed a problem where users were not able to save large Notebooks.
- Fixed a display issue when using JupyterLab's simple interface.
- Improved timeout behavior switch hardware operations.
- Improved error messaging when a service account cannot access the Runtime.
- Security fixes.
- Regular package refreshment and bug fixes.

[Learn more about managed notebooks versions](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/versions).
Fixed Fixed a server-side request forgery (SSRF) vulnerability. Previous versions of managed notebooks and user-managed notebooks instances still contain the vulnerability. It is recommended that you [migrate your data](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/migrate) to a new instance.
**Vertex AI** v1 Feature The option to configure pipeline run caching, ([`enable_caching`](https://docs.cloud.google.com/python/docs/reference/aiplatform/latest/google.cloud.aiplatform.PipelineJob.md.txt)), is available in the Google Cloud console.

## September 14, 2022

**Vertex AI** v1 Feature You can now limit the number of concurrent or parallel task runs in a pipeline run using `dsl.ParallelFor`. For more information, see the [Kubeflow Pipelines SDK Documentation](https://kubeflow-pipelines.readthedocs.io/en/sdk-2.0.1/source/dsl.html?h=dsl.parallelfor#).
Change The performance of the `ListPipelineJobs` API has been improved via a new [`readMask`](https://docs.cloud.google.com/vertex-ai/docs/reference/rest/v1/projects.locations.pipelineJobs/list#query-parameters) that lets you filter out large fields. To leverage this in the Python SDK, use the new [`enable_simple_view`](https://docs.cloud.google.com/python/docs/reference/aiplatform/latest/google.cloud.aiplatform.PipelineJob.md.txt#google_cloud_aiplatform_PipelineJob_list).

## August 17, 2022

**Agent Platform Workbench** v1beta1 \& v1 Feature **M95 release**

The M95 release of Vertex AI Workbench managed notebooks includes the following:

- Fixed a bug where users were regularly getting a 502 error when trying to access JupyterLab.
- Fixed a bug where opening an instance in Single User mode slowed the start of an instance.
- Fixed a bug where a managed notebooks instance was not starting after adding a GPU.
- Fixed bugs on the Serverless Spark form input.
- Improved the ActivityLog refresh after Serverless Spark creation.
- Fixed a bug related to the display of materialized views in BigQuery.
- Refreshed the JupyterLab interface with an improved Google-specific theme.
- Fixed a bug related to viewing Cloud Storage buckets and folders with large numbers of objects.
- Regular package refreshment and bug fixes.

[Learn more about managed notebooks versions](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/versions).

## August 12, 2022

**Vertex AI** v1 Feature **Vertex Explainable AI**

[Vertex Explainable AI](https://docs.cloud.google.com/vertex-ai/docs/explainable-ai/overview.md.txt) now offers [Preview](https://cloud.google.com/products/#product-launch-stages) support for example-based explanations. For more information, see [Configure example-based explanations for custom training](https://docs.cloud.google.com/vertex-ai/docs/explainable-ai/configuring-explanations-example-based.md.txt).

## August 01, 2022

**Vertex AI** v1 Feature **TensorFlow Profiler integration** : Debug model training performance for your custom training jobs. For details, see [Profile model training performance using Profiler](https://docs.cloud.google.com/vertex-ai/docs/experiments/tensorboard-profiler).

## July 29, 2022

**Vertex AI** v1 Feature Vertex AI now offers [Preview](https://cloud.google.com/products/#product-launch-stages) support for [Custom prediction routines (CPR)](https://docs.cloud.google.com/vertex-ai/docs/predictions/custom-prediction-routines). CPR lets you easily build custom containers for prediction with pre/post processing support.

## July 18, 2022

**Vertex AI** v1 Feature NFS support for custom training is GA. For details, see [Mount an NFS share for custom training](https://docs.cloud.google.com/vertex-ai/docs/training/train-nfs-share).

## July 14, 2022

**Vertex AI** v1 Feature The **Pipeline Templates** feature is available in [Preview](https://cloud.google.com/products#product-launch-stages). For documentation, refer to [Create, upload, and use a pipeline template](https://docs.cloud.google.com/vertex-ai/docs/pipelines/create-pipeline-template).

The features supported by pipeline templates include the following:

- Create a template registry using Artifact Registry (AR).
- Compile and publish a pipeline template.
- Create a pipeline run using the template and filter the runs.
- Manage (create, update, or delete) the pipeline template resources.

## July 12, 2022

**Vertex AI** v1 Change You can now use a pre-built container to perform custom training with [TensorFlow 2.9](https://docs.cloud.google.com/ai-platform-unified/docs/training/pre-built-containers#tensorflow)

## July 11, 2022

**Vertex AI** v1 Feature Vertex AI Pipelines now lets you configure task-level retries. You can set the number of times a task is retried before it fails. For more information about this option, see the [Kubeflow Pipelines SDK Documentation](https://kubeflow-pipelines.readthedocs.io/en/stable/source/dsl.html#kfp.dsl.PipelineTask.set_retry).

## July 06, 2022

**Vertex AI** v1 Feature **Tabular Workflows** is available in [Preview](https://cloud.google.com/products#product-launch-stages). For documentation, refer to [Tabular Workflows on Vertex AI](https://docs.cloud.google.com/vertex-ai/docs/tabular-data/tabular-workflows/overview).

**End-to-End AutoML** workflow is available in [Public Preview](https://cloud.google.com/products#product-launch-stages). For documentation, refer to [End-to-End AutoML](https://docs.cloud.google.com/vertex-ai/docs/tabular-data/tabular-workflows/e2e-automl).

## June 30, 2022

**Vertex AI** v1 Feature Feature: [Vertex AI Experiments](https://docs.cloud.google.com/vertex-ai/docs/experiments/intro-vertex-ai-experiments) is generally available ([GA](https://cloud.google.com/products#product-launch-stages)). Vertex AI Experiments helps users track and compare multiple experiment runs and analyze key model metrics.

Features supported by Experiments include:

- Vary and track parameters and metrics.
- Compare parameters, metrics, and artifacts between pipeline runs.
- Track steps and artifacts to capture the lineage of experiments.
- Compare vertex pipelines against Notebook experiments.

## June 28, 2022

**Vertex AI** v1 Feature **Vertex AI Forecasting** is available in [GA](https://cloud.google.com/products/#product-launch-stages). The following features are available:

- [Forecasting with ARIMA+](https://docs.cloud.google.com/vertex-ai/docs/tabular-data/forecasting-arima/overview)
- [Explainable AI for Forecasting](https://docs.cloud.google.com/vertex-ai/docs/tabular-data/forecasting-explanations)
- [Hierarchical Forecasting](https://docs.cloud.google.com/vertex-ai/docs/tabular-data/forecasting/hierarchical)
- [Model training with AutoML or Seq2Seq+](https://docs.cloud.google.com/vertex-ai/docs/tabular-data/forecasting/train-model)
- [Forecasting window configuration during model training](https://docs.cloud.google.com/vertex-ai/docs/tabular-data/forecasting/train-model)
- [Holiday effect modeling during model training](https://docs.cloud.google.com/vertex-ai/docs/tabular-data/forecasting/train-model)

## June 17, 2022

**Vertex AI** v1 Feature Support for IAM [resource-level policies](https://docs.cloud.google.com/vertex-ai/docs/featurestore/resource-policy.md.txt) for Vertex AI featurestore and entityType resources is available in [Preview](https://cloud.google.com/products#product-launch-stages).

## May 27, 2022

**Agent Platform Workbench** v1beta1 \& v1 Feature **M93 release**

The M93 release of Vertex AI Workbench managed notebooks includes the following:
Fixed
- Fixed a bug that prevented kernels from shutting down properly in Vertex AI Workbench managed notebooks.

[Learn more about managed notebooks versions](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/versions).

## May 24, 2022

**Vertex AI** v1 Feature You can now [configure the failure policy](https://docs.cloud.google.com/vertex-ai/docs/pipelines/configure-failure-policy) for a pipeline run.

## May 18, 2022

**Vertex AI** v1 Feature The ability to configure Vertex AI private endpoints is now general available ([GA](https://cloud.google.com/products#product-launch-stages)). Vertex AI private endpoints provide a low-latency, secure connection to the Vertex AI online prediction service. You can configure Vertex AI private endpoints by using VPC Network Peering. For more information, see [Use private endpoints for online prediction](https://docs.cloud.google.com/vertex-ai/docs/predictions/using-private-endpoints).

## May 12, 2022

**Agent Platform Workbench** v1beta1 \& v1 Feature **M91 release**

The M91 release of Vertex AI Workbench managed notebooks includes the following:

- Log streaming to the consumer project via Logs Viewer is now supported.
- Added the `net-tools` package.
- Regular package refreshments and bug fixes.
Fixed
- Fixed an issue that caused Spark server networking errors when using Dataproc Serverless Spark and VPC Peering.

[Learn more about managed notebooks versions](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/versions).

## April 26, 2022

**Vertex AI** v1 Feature You can now train your custom models using [Cloud TPU Architecture (TPU VMs)](https://docs.cloud.google.com/vertex-ai/docs/training/training-with-tpu-vm).

## April 21, 2022

**Vertex AI** v1 Feature You can now use a pre-built container to perform custom training with [PyTorch 1.11](https://docs.cloud.google.com/vertex-ai/docs/training/pre-built-containers#pytorch).

## April 06, 2022

**Agent Platform Workbench** v1beta1 \& v1 Feature [Vertex AI Workbench](https://docs.cloud.google.com/vertex-ai/docs/workbench) is generally available ([GA](https://cloud.google.com/products#product-launch-stages)). Vertex AI Workbench is a single notebook surface for all your data science needs that lets you access BigQuery data and Cloud Storage from within JupyterLab, execute notebook code in Vertex AI custom training and Spark, use custom containers, manage costs with idle timeout, and secure your instances with VPC Service Controls and customer managed encryption keys (CMEK).

Features supported include:

- Google-managed instances and the latest GPU support
- [Idle shutdown](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/idle-shutdown) for managed notebooks instances
- [Custom containers](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/custom-container)
- End-user and service account authentication
- Native plug-ins for [BigQuery](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/bigquery) and [Cloud Storage](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/cloud-storage)
- In-notebook Spark connect to Dataproc clusters
- Jobs support via the [managed notebooks executor](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/executor) on Vertex AI custom training and Spark
- One-click deploy for NGC containers
- VPC Service Controls
- [Customer managed encryption keys (CMEK)](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/cmek)
Feature The Vertex AI Workbench managed notebooks [executor](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/executor) is generally available ([GA](https://cloud.google.com/products#product-launch-stages)). Use the executor to run notebook files on a schedule or as a one-time execution. You can use parameters in your execution to make specific changes to each run. For example, you might specify a different dataset to use, change the learning rate on your model, or change the version of the model. For more information, see [Run notebook files with the executor](https://docs.cloud.google.com/vertex-ai/docs/workbench/managed/executor).
**Vertex AI** v1 Feature [Vertex AI Model Registry](http://cloud.google.com/vertex-ai/docs/model-registry/introduction) is available in [Preview](https://cloud.google.com/products/#product-launch-stages). Vertex AI Model Registry is a searchable repository where you can manage the lifecycle of your ML models. From the Vertex AI Model Registry, you can better organize your models, train new versions, and deploy directly to endpoints.

## March 07, 2022

**Vertex AI** v1 Feature [Google Cloud Pipeline Components SDK](https://docs.cloud.google.com/vertex-ai/docs/pipelines/components-introduction) v1.0 is now generally available.
Feature Vertex AI Feature Store [online store autoscaling](https://docs.cloud.google.com/vertex-ai/docs/featurestore/managing-featurestores.md.txt#scaling_options) is available in [Preview](https://cloud.google.com/products#product-launch-stages). The online store nodes automatically scale to balance performance and cost with different traffic patterns. The offline store already scales automatically.
Feature You can now mount Network File System (NFS)
shares to access remote files when you run a custom training job. For more information, see [Mount an NFS share for custom training](https://docs.cloud.google.com/vertex-ai/docs/training/train-nfs-share).

This feature is in [Preview](https://cloud.google.com/products#product-launch-stages).

## February 16, 2022

**Vertex AI** v1 Feature You can now use a pre-built container to perform custom training with [TensorFlow 2.8](https://docs.cloud.google.com/ai-platform-unified/docs/training/pre-built-containers#tensorflow).

## February 10, 2022

**Vertex AI** v1 Change For Vertex AI featurestore resources, the online store is optional. You can set the number of online nodes to `0`. For more information, see [Manage featurestores](https://docs.cloud.google.com/vertex-ai/docs/featurestore/managing-featurestores.md.txt).

## January 04, 2022

**Vertex AI** v1 Feature You can now use a pre-built container to perform custom training with [PyTorch 1.10](https://docs.cloud.google.com/vertex-ai/docs/training/pre-built-containers#pytorch).

## December 23, 2021

**Vertex AI** v1 Change There are now three Vertex AI release note feeds. Add any of the following to your [feed
reader](https://wikipedia.org/wiki/Comparison_of_feed_aggregators):

- For both Vertex AI and Vertex AI Workbench: `https://cloud.google.com/feeds/vertex-ai-product-group-release-notes.xml`
- For Vertex AI only: `https://cloud.google.com/feeds/vertex-ai-release-notes.xml`
- For Vertex AI Workbench only: `https://cloud.google.com/feeds/aiplatformnotebooks-release-notes.xml`

## December 02, 2021

**Vertex AI** v1 Feature You can now use a pre-built container to perform custom training with [TensorFlow 2.7](https://docs.cloud.google.com/ai-platform-unified/docs/training/pre-built-containers#tensorflow).

## December 01, 2021

**Vertex AI** v1 Feature [Vertex AI TensorBoard](https://docs.cloud.google.com/vertex-ai/docs/experiments/tensorboard-overview) is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

## November 19, 2021

**Vertex AI** v1 Feature The [autopackaging](https://docs.cloud.google.com/vertex-ai/docs/training/create-custom-job) feature of the `gcloud ai custom-jobs create` command is generally available ([GA](https://cloud.google.com/products#product-launch-stages)). Autopackaging lets you use a single command to run code on your local computer as a custom training job in Vertex AI.
Feature The `gcloud ai customs-jobs local-run` command is generally available ([GA](https://cloud.google.com/products#product-launch-stages)). You can use this command to [containerize and run training code locally](https://docs.cloud.google.com/vertex-ai/docs/training/containerize-run-code-local).

## November 09, 2021

**Vertex AI** v1 Feature [Vertex AI Pipelines](https://docs.cloud.google.com/vertex-ai/docs/pipelines/introduction) is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

## November 02, 2021

**Vertex AI** v1 Feature Using [interactive shells](https://docs.cloud.google.com/vertex-ai/docs/training/monitor-debug-interactive-shell) to inspect custom training jobs is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

You can use these interactive shells with [VPC Service Controls](https://docs.cloud.google.com/vertex-ai/docs/general/vpc-service-controls).

## October 25, 2021

**Vertex AI** v1 Feature [Vertex ML Metadata](https://docs.cloud.google.com/vertex-ai/docs/ml-metadata) is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

## October 11, 2021

**Agent Platform Workbench** v1beta1 \& v1 Feature [Vertex AI Workbench](https://docs.cloud.google.com/vertex-ai/docs/workbench) is now available in [Preview](https://cloud.google.com/products#product-launch-stages). Vertex AI Workbench is a notebook-based development environment for the entire data science workflow.
Announcement The Notebooks product and all existing Notebooks instances are now part of Vertex AI Workbench as [user-managed notebooks](https://docs.cloud.google.com/vertex-ai/docs/workbench/user-managed/introduction).

## October 05, 2021

**Vertex AI** v1 Feature [Vertex Feature Store](https://docs.cloud.google.com/vertex-ai/docs/featurestore/overview.md.txt) is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

## September 24, 2021

**Vertex AI** v1 Feature [Vertex Matching Engine](https://docs.cloud.google.com/vertex-ai/docs/matching-engine/overview) is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

## September 21, 2021

**Vertex AI** v1 Feature [Vertex AI Vizier](https://docs.cloud.google.com/vertex-ai/docs/vizier/overview) is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

## September 15, 2021

**Vertex AI** v1 Feature [Vertex Explainable AI](https://docs.cloud.google.com/vertex-ai/docs/explainable-ai/overview.md.txt) is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).

## September 13, 2021

**Vertex AI** v1 Feature You can use a [pre-built container to serve predictions from TensorFlow 2.6 models](https://docs.cloud.google.com/vertex-ai/docs/predictions/pre-built-containers#tensorflow).

## September 10, 2021

**Agent Platform Workbench** v1beta1 \& v1 Change Due to a recent change, the `iam.serviceAccounts.actAs` permission on the specified service account for the notebook instance is required for users to continue to have access to their notebook instances. The Google internal Inverting Proxy server that provides access to notebook instances now verifies that this permission is present before allowing users access to the JupyterLab URL. The JupyterLab URL this update covers is:
`*.notebooks.googleusercontent.com`
This update only applies to notebook instances in **Single User** mode and verifies that the assigned single user is authorized to execute code inside the notebook instance. Notebook instances running in Service Account or Project Editor mode already perform this verification via the Inverting Proxy server.
**Vertex AI** v1 Feature [Vertex Model Monitoring](https://docs.cloud.google.com/vertex-ai/docs/model-monitoring/overview) is generally available ([GA](https://cloud.google.com/products#product-launch-stages)).
Feature When you perform custom training, you can [access Cloud Storage buckets by reading and writing to the local filesystem](https://docs.cloud.google.com/vertex-ai/docs/training/code-requirements#fuse). This feature, based on [Cloud Storage Fuse](https://docs.cloud.google.com/storage/docs/gcs-fuse), is available in [Preview](https://cloud.google.com/products#product-launch-stages).

## August 30, 2021

**Vertex AI** v1 Feature You can now use a pre-built container to perform custom training with [TensorFlow 2.6](https://docs.cloud.google.com/ai-platform-unified/docs/training/pre-built-containers#tensorflow) and [PyTorch 1.9](https://docs.cloud.google.com/ai-platform-unified/docs/training/pre-built-containers#pytorch).

## August 24, 2021

**Vertex AI** v1 Feature The following tools for creating embeddings to use with [Vertex Matching Engine](https://docs.cloud.google.com/vertex-ai/docs/matching-engine/overview) are available in [Preview](https://cloud.google.com/products#product-launch-stages):

- the Two Tower built-in algorithm
- the Swivel pipeline template

## August 02, 2021

**Vertex AI** v1 Change Vertex Pipelines is available in the following regions:

- `us-east1` (South Carolina)
- `europe-west2` (London)
- `asia-southeast1` (Singapore)

See all the [locations where Vertex Pipelines is available](https://docs.cloud.google.com/vertex-ai/docs/general/locations).

## July 28, 2021

**Vertex AI** v1 Feature You can [use the Reduction Server algorithm](https://docs.cloud.google.com/vertex-ai/docs/training/distributed-training#reducing_training_time_with_reduction_server) ([Preview](https://cloud.google.com/products#product-launch-stages)) to increase throughput and reduce latency during distributed custom training.

## July 27, 2021

**Vertex AI** v1 Feature The following features are generally available ([GA](https://cloud.google.com/products#product-launch-stages)):

- [Access Transparency for Vertex AI](https://docs.cloud.google.com/vertex-ai/docs/general/access-transparency)
- [Using a custom service account for custom training and prediction](https://docs.cloud.google.com/vertex-ai/docs/general/custom-service-account)
- [Using VPC Service Controls with Vertex AI](https://docs.cloud.google.com/vertex-ai/docs/general/vpc-service-controls)
- [Setting up VPC Network Peering with Vertex AI](https://docs.cloud.google.com/vertex-ai/docs/general/vpc-peering) and [using private IP for custom training](https://docs.cloud.google.com/vertex-ai/docs/training/using-private-ip) (Using [private IP for prediction](https://docs.cloud.google.com/vertex-ai/docs/predictions/using-private-endpoints) and [vector matching with Matching Engine](https://docs.cloud.google.com/vertex-ai/docs/matching-engine/using-matching-engine) remains in preview.)

## July 26, 2021

**Agent Platform Workbench** v1beta1 \& v1 Change If using proxy single-user mode, Notebooks API now verifies if the specified user (`proxy-user-mail`) has Service Account permissions on the Service Account. This check is performed during instance creation and registration.

## July 20, 2021

**Vertex AI** v1 Feature Private endpoints for online prediction are now available in preview.
After you [set up VPC Network Peering with Vertex AI](https://docs.cloud.google.com/vertex-ai/docs/general/vpc-peering), you can [create private endpoints for low-latency online prediction](https://docs.cloud.google.com/vertex-ai/docs/predictions/using-private-endpoints) within your private network.

Additionally, the documentation for VPC Network Peering with custom training has moved. The general instructions for setting up VPC Network Peering with Vertex AI are available at the original link, https://cloud.google.com/vertex-ai/docs/general/vpc-peering. The documentation for custom training is now available here: [Using private IP with custom training](https://docs.cloud.google.com/vertex-ai/docs/training/using-private-ip).

## July 19, 2021

**Vertex AI** v1 Feature You can now [use an interactive shell](https://docs.cloud.google.com/vertex-ai/docs/training/monitor-debug-interactive-shell) to inspect your custom training container while it runs. The interactive shell can be helpful for monitoring and debugging training.

This feature is available in preview.

## July 14, 2021

**Vertex AI** v1 Feature You can now [use the `gcloud beta ai custom-jobs create` command](https://docs.cloud.google.com/vertex-ai/docs/training/create-custom-job#create) to build a Docker image based on local training code, push the image to Container Registry, and create a `CustomJob` resource.

## July 08, 2021

**Vertex AI** v1 Feature You can now [containerize and run your training code locally](https://docs.cloud.google.com/vertex-ai/docs/training/containerize-run-code-local) by using the new `gcloud beta ai custom-jobs local-run` command. This feature is available in preview.

## June 25, 2021

**Vertex AI** v1 Feature You can now use [NVIDIA A100 GPUs](https://cloud.google.com/vertex-ai/docs/training/configure-compute#specifying_gpus) and several [accelerator-optimized (A2) machine types](https://docs.cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types) for training. You must use A100 GPUs and A2 machine types together. Learn about [their pricing](https://docs.cloud.google.com/vertex-ai/pricing).

## June 18, 2021

**Agent Platform Workbench** v1 Feature Support for Compute Reservations. Notebooks API allows the use of Compute Reservations during instance creation.

## June 11, 2021

**Vertex AI** v1 Feature You can now use a [pre-built container to serve predictions from TensorFlow 2.5 models](https://docs.cloud.google.com/vertex-ai/docs/predictions/pre-built-containers#tensorflow).
Feature You can now use a [pre-built container to serve predictions from XGBoost 1.4 models](https://docs.cloud.google.com/vertex-ai/docs/predictions/pre-built-containers#xgboost).

## May 18, 2021

**Vertex AI** v1 Announcement AI Platform (Unified) is now [Vertex AI](https://docs.cloud.google.com/vertex-ai).
Feature Vertex AI now supports [forecasting with time series data](https://docs.cloud.google.com/vertex-ai/docs/start/automl-model-types#tabular) for AutoML tabular models, in [Preview](https://cloud.google.com/products#product-launch-stages). You can use forecasting to predict a series of numeric values that extend into the future.
Feature Vertex AI has added support for custom model training, custom model batch prediction, custom model online prediction, and a limited number of other services in the following [regions](https://docs.cloud.google.com/vertex-ai/docs/general/locations):

- us-west1
- us-east1
- us-east4
- northamerica-northeast1
- europe-west2
- europe-west1
- asia-southeast1
- asia-northeast1
- australia-southeast1
- asia-northeast3
Feature [Vertex Pipelines](https://docs.cloud.google.com/vertex-ai/docs/pipelines/introduction) is now available in [Preview](https://cloud.google.com/products#product-launch-stages). Vertex Pipelines helps you to automate, monitor, and govern your ML systems by orchestrating your ML workflow.
Feature [Vertex Matching Engine](https://docs.cloud.google.com/vertex-ai/docs/matching-engine/overview) is now available in [Preview](https://cloud.google.com/products#product-launch-stages). Vertex Matching Engine enables vector similarity search.
Feature [Vertex Model Monitoring](https://docs.cloud.google.com/vertex-ai/docs/model-monitoring/overview) is now available in [Preview](https://cloud.google.com/products#product-launch-stages). Vertex Model Monitoring enables you to monitor model quality over time.
Feature [Vertex Feature Store](https://docs.cloud.google.com/vertex-ai/docs/featurestore/overview.md.txt) is now available in [Preview](https://cloud.google.com/products#product-launch-stages). Vertex Feature Store provides a centralized repository for organizing, storing, and serving ML features.
Feature [Vertex TensorBoard](https://docs.cloud.google.com/vertex-ai/docs/experiments/tensorboard-overview) is now available in [Preview](https://cloud.google.com/products#product-launch-stages). Vertex TensorBoard enables you to track, visualize, and compare ML experiments.
Feature [Vertex ML Metadata](https://docs.cloud.google.com/vertex-ai/docs/ml-metadata/introduction) is now available in [Preview](https://cloud.google.com/products#product-launch-stages). Vertex ML Metadata lets you record the metadata and artifacts produced by your ML system so you can analyze the performance of your ML system.

## May 03, 2021

**Vertex AI** v1 Feature You can now use a [pre-built container to serve predictions from XGBoost 1.3 models](https://docs.cloud.google.com/ai-platform-unified/docs/predictions/pre-built-containers#xgboost).
Feature You can now use a [pre-built container to serve predictions from scikit-learn 0.24 models](https://docs.cloud.google.com/ai-platform-unified/docs/predictions/pre-built-containers#scikit-learn).
Feature You can now use a [pre-built container to serve predictions from TensorFlow 2.4 models](https://docs.cloud.google.com/ai-platform-unified/docs/predictions/pre-built-containers#tensorflow).

## April 27, 2021

**Vertex AI** v1 Feature [AI Platform Vizier](https://docs.cloud.google.com/ai-platform-unified/docs/vizier/overview) is now available in preview. Vizier is a feature of AI Platform (Unified) that you can use to perform black-box optimization. You can use Vizier to tune hyperparameters or optimize any evaluable system.

## April 15, 2021

**Vertex AI** v1 Feature The Python client library for AI Platform (Unified) is now called the
AI Platform (Unified) SDK. With the release of version 0.7
([Preview](https://github.com/googleapis/google-cloud-python/blob/master/README.rst#beta-support)), the AI Platform (Unified) SDK provides two levels of support.
The high-level `aiplatform` library
is designed to simplify common data
science workflows by using wrapper classes and opinionated defaults. The
lower-level `aiplatform.gapic` library remains
available for those times when you need more flexibility or control.
[Learn more](https://docs.cloud.google.com/ai-platform-unified/docs/start/client-libraries).

## March 31, 2021

**Vertex AI** v1 Announcement AI Platform (Unified) is now available in General Availability (GA).
Change AI Platform (Unified) has added support for the following regions for custom model training, as well as batch and online prediction for custom-trained models:

- us-west1 (Oregon)
- us-east1 (South Carolina)
- us-east4 (N. Virginia)
- northamerica-northeast1 (Montreal)
- europe-west2 (London)
- europe-west1 (Belgium)
- asia-southeast1 (Singapore)
- asia-northeast1 (Tokyo)
- australia-southeast1 (Sydney)
- asia-northeast3 (Seoul)

## March 26, 2021

**Agent Platform Workbench** v1 Feature Cross Project Service Account is supported for user-managed notebooks.

## March 15, 2021

**Vertex AI** v1 Feature You can now use a [pre-built container to perform custom training with PyTorch 1.7](https://docs.cloud.google.com/ai-platform-unified/docs/training/pre-built-containers#pytorch).

## March 04, 2021

**Agent Platform Workbench** v1 Change New Notebooks instances add labels for VM image (`goog-caip-notebook`) and volume (`goog-caip-notebook-volume`).

## March 02, 2021

**Vertex AI** v1 Change **CMEK compliance using the client libraries**

You can now use the client libraries to create resources with a customer-managed encryption key (CMEK).

For more information on creating a resource with an encryption key using the client libraries, see [Using customer-managed encryption keys (CMEK)](https://docs.cloud.google.com/ai-platform-unified/docs/general/cmek#create_resources_with_the_kms_key).

## March 01, 2021

**Vertex AI** v1 Change The [client library](https://docs.cloud.google.com/ai-platform-unified/docs/start/client-libraries) for Java now includes enhancements to improve usage of training and prediction features. The client library includes additional types and utility functions for sending training requests, sending prediction requests, and reading prediction results.

To use these enhancements, you must install the latest version of the [client library](https://docs.cloud.google.com/ai-platform-unified/docs/start/client-libraries).

## February 25, 2021

**Vertex AI** v1 Feature AI Platform (Unified) now supports Access Transparency in beta. Google Cloud organizations with certain support packages can use this feature. Learn more about [using Access Transparency with AI Platform (Unified)](https://docs.cloud.google.com/ai-platform-unified/docs/general/access-transparency).
Change The [client libraries](https://docs.cloud.google.com/ai-platform-unified/docs/start/client-libraries) for Node.js and Python now include enhancements to improve usage of training and prediction features. These client libraries include additional types and utility functions for sending training requests, sending prediction requests, and reading prediction results.

To use these enhancements, you must install the latest version of the
[client libraries](https://docs.cloud.google.com/ai-platform-unified/docs/start/client-libraries).
Change The `predict` and `explain` method calls no longer require the use of a different service endpoint (for example, `https://us-central1-prediction-aiplatform.googleapis.com`). These methods are now available on the same endpoint as all other methods.
Feature In addition to Docker images hosted on [Container Registry](https://docs.cloud.google.com/container-registry/docs), you can now use Docker images hosted on [Artifact Registry](https://docs.cloud.google.com/artifact-registry/docs.md.txt) and [Docker Hub](https://hub.docker.com/) for [custom container training on AI Platform](https://docs.cloud.google.com/ai-platform-unified/docs/training/containers-overview).
Change The Docker images for [pre-built training containers](https://docs.cloud.google.com/ai-platform-unified/docs/training/pre-built-containers) and [pre-built prediction containers](https://docs.cloud.google.com/ai-platform-unified/docs/predictions/pre-built-containers) are now available on Artifact Registry.
Feature You can now use a [pre-built container to serve predictions from XGBoost 1.2 models](https://docs.cloud.google.com/ai-platform-unified/docs/predictions/pre-built-containers#xgboost).
Feature You can now use a [pre-built container to perform custom training with TensorFlow 2.4](https://docs.cloud.google.com/ai-platform-unified/docs/training/pre-built-containers#tensorflow).
Feature You can now use a [pre-built container to serve predictions from TensorFlow 2.3 models](https://docs.cloud.google.com/ai-platform-unified/docs/predictions/pre-built-containers#tensorflow).

## February 01, 2021

**Agent Platform Workbench** v1 Feature Notebooks [Terraform Module](https://registry.terraform.io/providers/hashicorp/google/latest/docs/resources/notebooks_instance) supports Notebooks API v1
**Vertex AI** v1 Feature You can now use a [pre-built container to perform custom training with PyTorch 1.6](https://docs.cloud.google.com/ai-platform-unified/docs/training/pre-built-containers#pytorch).

## January 23, 2021

**Agent Platform Workbench** v1 Feature Notebooks API supports [Shielded VM configuration](https://cloud.google.com/security/products/shielded-vm).
Announcement VPC-SC for Notebooks (now known as user-managed notebooks) is now [Generally Available](https://cloud.google.com/products#product-launch-stages).

## January 19, 2021

**Vertex AI** v1 Feature **Preview** : Select AI Platform (Unified) resources can now be configured to use [Customer-managed encryption keys (CMEK)](https://docs.cloud.google.com/ai-platform-unified/docs/general/cmek).

Currently you can only create resources with a CMEK key in the UI; this functionality is not currently available using the client libraries.

## January 11, 2021

**Vertex AI** v1 Breaking The default boot disk type for virtual machine instances used for custom training has changed from `pd-standard` to `pd-ssd`. Learn more about [disk types for custom training](https://cloud.google.com/ai-platform-unified/docs/training/configure-compute#specifying_boot_disk_options) and read about [pricing for different disk types](https://docs.cloud.google.com/ai-platform-unified/pricing#training).

If you previously used the default disk type for custom training and want to continue training with the same disk type, make sure to explicitly specify the `pd-standard` boot disk type when you perform custom training.

## January 06, 2021

**Vertex AI** v1 Feature You can now use a [pre-built container to perform custom training with TensorFlow 2.3](https://docs.cloud.google.com/ai-platform-unified/docs/training/pre-built-containers#tensorflow).

## December 17, 2020

**Vertex AI** v1 Feature AI Platform (Unified) now stores and
processes your data only in the region you specify for most features.
[Learn more](https://docs.cloud.google.com/ai-platform-unified/docs/general/locations).

## November 16, 2020

**Vertex AI** v1 Feature **Preview release**

AI Platform (Unified) is now available in [Preview](https://cloud.google.com/products#product-launch-stages).

For more information, see the [product documentation](https://docs.cloud.google.com/ai-platform-unified/docs).

## September 21, 2020

**Agent Platform Workbench** v1 Feature AI Platform Notebooks (now known as user-managed notebooks) API is now [Generally Available](https://cloud.google.com/products#product-launch-stages). The API now includes an [isUpgradable](https://docs.cloud.google.com/ai-platform/notebooks/docs/reference/rest/v1/projects.locations.instances/isUpgradeable) endpoint and adds manual and auto-upgrade functionality to notebooks instances created using the API.
Feature [Cloud Audit Logging for AI Platform Notebooks](https://docs.cloud.google.com/vertex-ai/docs/workbench/user-managed/audit-logging) (now known as user-managed notebooks) is now [Generally Available](https://cloud.google.com/products#product-launch-stages).

[Granular IAM permissions for AI Platform Notebooks](https://docs.cloud.google.com/vertex-ai/docs/workbench/user-managed/iam) (now known as user-managed notebooks) is now [Generally Available](https://cloud.google.com/products#product-launch-stages).

AI Platform Notebooks now supports [E2 machine types](https://docs.cloud.google.com/compute/docs/machine-types#e2_machine_types).

The following new regions have been added:

- `europe-west2` (London, UK)
- `europe-west3` (Frankfurt, Germany)
- `europe-west6` (Zürich, Switzerland)

## March 31, 2020

**Agent Platform Workbench** v1beta1 Feature AI Platform Notebooks (now known as user-managed notebooks) is now [Generally Available](https://cloud.google.com/products#product-launch-stages). Some integrations with and specific features of AI Platform Notebooks are still in beta, such as [Virtual Private Cloud Service Controls](https://docs.cloud.google.com/ai-platform/notebooks/docs/service-perimeter), [Identity and Access Management (IAM)](https://docs.cloud.google.com/ai-platform/notebooks/docs/iam) roles, and [AI Platform Notebooks API](https://docs.cloud.google.com/ai-platform/notebooks/docs/reference/rest).

## February 04, 2020

**Agent Platform Workbench** v1beta1 Feature [VPC Service Controls](https://docs.cloud.google.com/vpc-service-controls/docs/overview.md.txt) now supports AI Platform Notebooks. Learn [how to use a notebook instance within a service perimeter](https://docs.cloud.google.com/ai-platform/notebooks/docs/service-perimeter). This functionality is in [beta](https://cloud.google.com/products/#product-launch-stages).

## February 03, 2020

**Agent Platform Workbench** v1beta1 Feature AI Platform Notebooks now supports Access Transparency. Access Transparency
provides you with logs of actions that Google staff have taken when accessing
your data. To learn more about Access Transparency, see the [Overview of
Access Transparency](https://docs.cloud.google.com/assured-workloads/access-transparency/docs/overview).

## September 12, 2019

**Agent Platform Workbench** v1 Feature You can now use customer-managed encryption keys (CMEK) to protect data on the boot disks of your AI Platform Notebooks (now known as user-managed notebooks) VM instances. CMEK in AI Platform Notebooks is generally available. For more information, see [Using customer-managed encryption keys (CMEK)](https://docs.cloud.google.com/vertex-ai/docs/workbench/user-managed/cmek).

## September 09, 2019

**Agent Platform Workbench** v1beta1 Feature AI Platform Notebooks now provides more ways for you to customize your network settings, encrypt your notebook content, and grant access to your notebook instance. These options are available when you [create a notebook](https://docs.cloud.google.com/ai-platform/notebooks/docs/create-new#create-with-options).
Feature Now you can implement AI Platform Notebooks using custom containers. Use a [Deep Learning Containers image](https://docs.cloud.google.com/ai-platform/deep-learning-containers/docs/choosing-container#choose_a_container_image_type) or [create a derivative container](https://docs.cloud.google.com/ai-platform/deep-learning-containers/docs/derivative-container) of your own, then [create a new notebook instance using your custom container](https://docs.cloud.google.com/ai-platform/notebooks/docs/custom-container).

## July 12, 2019

**Agent Platform Workbench** v1beta1 Change R upgraded to version 3.6.

R Notebooks are no longer dependent on a Conda environment.

## June 03, 2019

**Agent Platform Workbench** v1beta1 Feature You can now create AI Platform Notebooks instances with R and core R packages installed. Learn [how to install R dependencies](https://docs.cloud.google.com/ai-platform/notebooks/docs/dependencies), and read guides for [using R with BigQuery in AI Platform Notebooks](https://docs.cloud.google.com/ai-platform/notebooks/docs/use-r-bigquery) and [using R and Python in the same notebook](https://docs.cloud.google.com/ai-platform/notebooks/docs/r-python-same-notebook).

## March 01, 2019

**Agent Platform Workbench** v1beta1 Feature [AI Platform Notebooks](https://docs.cloud.google.com/ai-platform/notebooks/docs/overview) is now available in beta. AI Platform Notebooks enables you to create and manage virtual machine (VM) instances that are pre-packaged with [JupyterLab](https://jupyterlab.readthedocs.io/en/stable/getting_started/overview.html) and a suite of deep learning software.

Visit the [AI Platform Notebooks overview](https://docs.cloud.google.com/ai-platform/notebooks/docs/overview) and the [guide to creating a new notebook instance](https://docs.cloud.google.com/ai-platform/notebooks/docs/create-new) to learn more.