Serverless mode

Preview

Some of the RAG features are Preview offerings, subject to the "Pre-GA Offerings Terms" of the Google Cloud Service Specific Terms. Pre-GA products and features are available "as-is" and may have limited support, and changes to Pre-GA products and features may not be compatible with other Pre-GA versions. For more information, see the launch stage descriptions. By using the Gemini API on Vertex AI, you agree to the Generative AI Preview terms and conditions (Preview Terms).

The VPC-SC security controls and CMEK are supported by Vertex AI RAG Engine. Data residency and AXT security controls aren't supported.

Serverless mode is the recommended way to get started with Vertex AI RAG Engine. It provides a fully-managed and highly scalable database to back your Vertex AI RAG Engine resources, without the hassle of provisioning or managing any infrastructure configurations. It offers quick onboarding and is suitable for most users, including enterprise scale.

Serverless mode is a preview feature and is available in us-central1 only. It is available to everyone and does not require any allowlist.

With Serverless mode, your Vertex AI RAG Engine CUJ remains identical to that with Spanner mode, except for the following key differences:

1. RagManagedDb is not available as a choice of vector database when using Serverless mode. Instead, the new default vector database is the fully-managed and highly scalable RagManagedVertexVectorSearch. To learn more about this vector DB, refer to Use Vector Search 2.0 with RAG.
2. Customer-Managed Encryption Keys (CMEK) are not supported in Serverless mode.

Switching to Serverless mode

To switch to Serverless mode as your choice of deployment mode, refer to the Switching your modes page.

Delete your data

To delete your data from Serverless deployment mode, follow these steps:

1. Ensure your active mode is set to Serverless.
2. Call the ListRagCorpora API to view your resources under this mode,
3. Delete each corpus using the DeleteRagCorpus API.

Pricing

There is no additional charge when using Serverless deployment mode. However, if you have an active Spanner mode, you might still get charged for it. Additionally, you might incur charges from your choice of models, reranking, and vector storage on Vertex AI RAG Engine. For more information, refer to the Vertex AI RAG Engine billing page.