Model Serving limits and regions

Preview

Mosaic AI Model Serving is in Public Preview and is supported in us-east1 and us-central1.

This article summarizes the limitations and region availability for Mosaic AI Model Serving and supported endpoint types.

Resource and payload limits

Mosaic AI Model Serving imposes default limits to ensure reliable performance. If you have feedback on these limits, reach out to your Databricks account team.

The following table summarizes resource and payload limitations for model serving endpoints.

Feature	Granularity	Limit
Payload size	Per request	16 MB. For endpoints serving external models the limit is 4 MB.
Queries per second (QPS)	Per workspace	200, but can be increased to 25,000 or more by reaching out to your Databricks account team.
Model execution duration	Per request	120 seconds
CPU endpoint model memory usage	Per endpoint	4GB
Provisioned concurrency	Per workspace	200 concurrency. Can be increased by reaching out to your Databricks account team.
Overhead latency	Per request	Less than 50 milliseconds
Init scripts		Init scripts are not supported.
Foundation Model APIs (pay-per-token) rate limits	Per workspace	Llama 3.3 70B Instruct has a limit of 2 queries per second and 1200 queries per hour. If this limit is are insufficient for your use case, Databricks recommends using provisioned throughput.
Foundation Model APIs (provisioned throughput) rate limits	Per workspace	200

Networking and security limitations

• Model Serving endpoints are protected by access control and respect networking-related ingress rules configured on the workspace.

• Model Serving does not provide security patches to existing model images because of the risk of destabilization to production deployments. A new model image created from a new model version will contain the latest patches. Reach out to your Databricks account team for more information.

Foundation Model APIs limits

Note

As part of providing the Foundation Model APIs, Databricks might process your data outside of the region and cloud provider where your data originated.

For both pay-per-token and provisioned throughput workloads:

Only workspace admins can change the governance settings, such as rate limits for Foundation Model APIs endpoints. To change rate limits, use the following steps:

1. Open the Serving UI in your workspace to see your serving endpoints.

2. From the kebab menu on the Foundation Model APIs endpoint you want to edit, select View details.

3. From the kebab menu on the upper-right side of the endpoints details page, select Change rate limit.

Pay-per-token limits

The following are limits relevant to Foundation Model APIs pay-per-token workloads:

• Pay-per-token workloads are not HIPAA or compliance security profile compliant.

• Meta Llama 3.3 70B Instruct is only available in pay-per-token US supported regions.

Provisioned throughput limits

The following are limits relevant to Foundation Model APIs provisioned throughput workloads:

• Provisioned throughput supports the HIPAA compliance profile and is recommended for workloads that require compliance certifications.

  • The GTE Large (En) embedding models do not generate normalized embeddings.

  • The following table shows the region availability of the supported Meta Llama 3.1, 3.2 and 3.3 models. See Deploy fine-tuned foundation models for guidance on how to deploy fine-tuned models.

  Meta Llama model variant	Regions
  meta-llama/Llama-3.1-8B-Instruct	us-east1 us-central1
  meta-llama/Llama-3.1-70B-Instruct	us-east1 us-central1
  meta-llama/Llama-3.2-1B-Instruct	us-east1 us-central1
  meta-llama/Llama-3.2-3B-Instruct	us-east1 us-central1

Region availability

Note

If you require an endpoint in an unsupported region, reach out to your Databricks account team.

If your workspace is deployed in a region that supports model serving but is served by a control plane in an unsupported region, the workspace does not support model serving. If you attempt to use model serving in such a workspace, you will see in an error message stating that your workspace is not supported. Reach out to your Databricks account team for more information.

For more information on regional availability of features, see Model serving regional availability.