Model Armor feature availability varies by region due to
dependencies on services that could process data outside the location. As a
result, some regions have limited feature support when you configure
Model Armor using templates in order to maintain strict
data residency. This page provides a breakdown of
supported features across all available regions.

> [!NOTE]
> **Note:** For Model Armor configurations that use floor settings, all features are available, but data residency enforcement for data in use and in transit varies by region.

## Full-support regions

If you create a Model Armor template in a full-support region,
you have access to all of the following filters. All underlying dependency
services are available in the jurisdictional boundary.

- [Prompt injection and
  jailbreak detection](https://docs.cloud.google.com/model-armor/overview#ma-prompt-injection)
- [Responsible AI](https://docs.cloud.google.com/model-armor/overview#ma-responsible-ai-safety-categories)
- [Sensitive Data Protection](https://docs.cloud.google.com/model-armor/overview#ma-sensitive-data-prot)
- [Malicious URL detection](https://docs.cloud.google.com/model-armor/overview#ma-malicious-url-detection)
- [Child sexual abuse material (CSAM) detection](https://docs.cloud.google.com/model-armor/overview#ma-csam-detection)
- [Multi-language detection](https://docs.cloud.google.com/model-armor/overview#languages-supported)
- Antivirus scanning

The following table shows the locations with full feature support.
The **Jurisdiction** column indicates the geographical boundary (for example, a
country or a multi-region like `eu` or `us`) within which
Model Armor ensures data residency compliance for the supported
states (at rest, in use, and in transit) as indicated in the table.

The **Region/multi-region** column lists the specific Google Cloud region or
multi-region identifier.

When data residency is enforced for a region in Model Armor, it
helps ensure that data remains within a specified jurisdiction while in at least
one of the following states:

- **At rest**: Stored data remains within the specific Google Cloud region listed.

- **In use** and **In transit** : Data residency enforcement for these states
  within the specified jurisdiction depends on the values in the **In use** and
  **In transit** columns:

  - **Yes** : Data residency is enforced for the *In use* and *In transit* states within the listed jurisdiction. This means that the data remains within the boundaries of the listed jurisdiction while being processed or transmitted (not necessarily within that single region).
  - **No** : Data residency is not enforced for the *In use* and *In transit* states within the listed jurisdiction. This means that the data might be processed or transmitted outside of the jurisdiction.

> [!NOTE]
> **Note:** For Assured Workloads, Model Armor maintains data residency within the selected jurisdiction for [supported locations](https://docs.cloud.google.com/model-armor/locations) for data in use. For more information, see [Assured Workloads Service Specific Terms](https://cloud.google.com/terms/service-terms#37).

| Region/multi-region | Jurisdiction | At rest | In use | In transit |
|---|---|---|---|---|
| `eu` | [European Union](https://docs.cloud.google.com/assured-workloads/docs/control-packages/eu-data-boundary) | Yes | Yes | Yes |
| `europe-southwest1` | [European Union](https://docs.cloud.google.com/assured-workloads/docs/control-packages/eu-data-boundary) | Yes | Yes | Yes |
| `europe-west1` | [European Union](https://docs.cloud.google.com/assured-workloads/docs/control-packages/eu-data-boundary) | Yes | Yes | Yes |
| `europe-west3` | [European Union](https://docs.cloud.google.com/assured-workloads/docs/control-packages/eu-data-boundary) | Yes | Yes | Yes |
| `europe-west4` | [European Union](https://docs.cloud.google.com/assured-workloads/docs/control-packages/eu-data-boundary) | Yes | Yes | Yes |
| `europe-west9` | [European Union](https://docs.cloud.google.com/assured-workloads/docs/control-packages/eu-data-boundary) | Yes | Yes | Yes |
| `us` | [United States](https://docs.cloud.google.com/assured-workloads/docs/control-packages/us-data-boundary) | Yes | Yes | Yes |
| `us-central1` | [United States](https://docs.cloud.google.com/assured-workloads/docs/control-packages/us-data-boundary) | Yes | Yes | Yes |
| `us-east1` | [United States](https://docs.cloud.google.com/assured-workloads/docs/control-packages/us-data-boundary) | Yes | Yes | Yes |
| `us-east4` | [United States](https://docs.cloud.google.com/assured-workloads/docs/control-packages/us-data-boundary) | Yes | Yes | Yes |
| `us-west1` | [United States](https://docs.cloud.google.com/assured-workloads/docs/control-packages/us-data-boundary) | Yes | Yes | Yes |

<br />

For a list of features that are available for each region or multi-region, see
[Supported features by region](https://docs.cloud.google.com/model-armor/feature-availability-by-region#supported-by-region) in this document.

## Limited-support regions

By default, if you create a Model Armor template in a
limited-support region and your template has [data residency compliance
enabled](https://docs.cloud.google.com/model-armor/manage-templates#set-data-residency-compliance), some
Model Armor features are unavailable because they rely on
services that might process data outside the [jurisdiction of your chosen
Model Armor
region](https://docs.cloud.google.com/model-armor/data-residency#data-residency-enforced).

If you must use a limited-support region but you want to prioritize
comprehensive security filtering over strict data residency, you can [disable
data residency
enforcement](https://docs.cloud.google.com/model-armor/manage-templates#set-data-residency-compliance) for
in-use and in-transit data in your Model Armor template.

The following are the regions with limited feature support:

| Region/multi-region | Jurisdiction | At rest | In use | In transit |
|---|---|---|---|---|
| `asia-northeast1` | [Japan](https://docs.cloud.google.com/assured-workloads/docs/control-packages/japan-data-boundary) | Yes | Yes | Yes |
| `asia-northeast3` | [South Korea](https://docs.cloud.google.com/assured-workloads/docs/control-packages/south-korea-data-boundary) | Yes | Yes | Yes |
| `asia-south1` | [India](https://docs.cloud.google.com/assured-workloads/docs/control-packages/india-data-boundary) | Yes | Yes | Yes |
| `asia-southeast1` | [Singapore](https://docs.cloud.google.com/assured-workloads/docs/control-packages/singapore-data-boundary) | Yes | Yes | Yes |
| `australia-southeast2` | [Australia](https://docs.cloud.google.com/assured-workloads/docs/control-packages/australia-data-boundary) | Yes | Yes | Yes |
| `europe-west2` | [United Kingdom](https://docs.cloud.google.com/assured-workloads/docs/control-packages/uk-data-boundary) | Yes | Yes | Yes |
| `northamerica-northeast2` | [Canada](https://docs.cloud.google.com/assured-workloads/docs/control-packages/canada-data-boundary) | Yes | Yes | Yes |
| `southamerica-east1` | [Brazil](https://docs.cloud.google.com/assured-workloads/docs/control-packages/brazil-data-boundary) | Yes | Yes | Yes |

For a list of features that are available for each region or multi-region, see
[Supported features by region](https://docs.cloud.google.com/model-armor/feature-availability-by-region#supported-by-region) in this document.

## Supported features by region

This section shows the regional capabilities and supported filters for each
region or multi-region when you configure Model Armor using a
template that has [data residency enforcement
enabled](https://docs.cloud.google.com/model-armor/manage-templates#set-data-residency-compliance).

| Region/multi-region | Supported filters | Multi-language detection | Image support | Antivirus scanning |
|---|---|---|---|---|
| `asia-northeast1` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak | No | No | No |
| `asia-northeast3` | - Sensitive Data Protection | No | No | No |
| `asia-south1` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak | No | No | No |
| `asia-southeast1` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak | No | No | No |
| `australia-southeast2` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak | No | No | No |
| `eu` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak - Malicious URL - CSAM - topicality | Yes | Yes | Yes |
| `europe-southwest1` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak - Malicious URL - CSAM - topicality | Yes | No | Yes |
| `europe-west1` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak - Malicious URL - CSAM - topicality | Yes | No | Yes |
| `europe-west2` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak | No | No | No |
| `europe-west3` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak - Malicious URL - CSAM - topicality | Yes | No | Yes |
| `europe-west4` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak - Malicious URL - CSAM - topicality | Yes | No | Yes |
| `europe-west9` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak - Malicious URL - CSAM - topicality | Yes | No | Yes |
| `northamerica-northeast2` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak | No | No | No |
| `southamerica-east1` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak | No | No | No |
| `us` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak - Malicious URL - CSAM - topicality | Yes | Yes | Yes |
| `us-central1` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak - Malicious URL | Yes | No | Yes |
| `us-east1` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak - Malicious URL - CSAM - topicality | Yes | No | Yes |
| `us-east4` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak - Malicious URL | Yes | No | Yes |
| `us-east7` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak - Malicious URL - CSAM - topicality | Yes | No | Yes |
| `us-west1` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak - Malicious URL - CSAM - topicality | Yes | No | Yes |
| `global` | - Responsible AI - Sensitive Data Protection - Prompt injection and jailbreak - Malicious URL - CSAM - topicality | Yes | Yes | Yes |