Request to Enable Foundation Model Endpoint Rate Limits

[KarthikRK]

Hi,

I'm working on a trade surveillance agent and need access to foundation model endpoints.
Currently getting this error when calling any LLM endpoint:

Error: "The endpoint is temporarily disabled due to a Databricks-set rate limit of 0"

Could you please enable rate limits (set to > 0) for at least ONE of these endpoints:
  • databricks-claude-sonnet-5 (preferred)
  • databricks-gpt-5-6-luna (alternative #1)

  • databricks-llama-4-maverick (alternative #2)

   

Workspace: https://dbc-ebe07208-acbf.cloud.databricks.com
Service Principal: f67f36e5-b977-4565-90f6-7c017d6f467a
Use Case: Trade surveillance agent with MCP tools for compliance

[balajij8 · Accepted solution]

What you are encountering on the free edition is the expected behavior. Databricks imposes a hard limit of 0 on premium, high demand Foundation Models especially - Claude and GPT groups when a workspace is operating under a free edition.

You can use llama 4 maverick or other Llama models without issues generally in the free tier. They are exempt from this specific zero-rate-limit hold, so you can swap the endpoint configuration to the Maverick model. You can reach out to your Databricks account team for the upgrade if you are already on a paid workspace and seeing this error

Free edition may not be used for commercial purposes. More details here

[KarthikRK]

llama 4 maverick worked to some extent but it's not fulfilling the expectations of my app; let me try other llama models and see

[KarthikRK]

@balajij8 - llama models stopped working today from workspace. What alternatives do we have ?

[balajij8]

@KarthikRK Free edition may not be used for commercial purposes. You can create a new Databricks account by starting a free trial or create a paid workspace if you would like access to key features like best models.

[KarthikRK]

@balajij8 -- 

qwen3-next-80b-a3b-instruct model has been very helpful for my POC/use case as it gives consistent and reproducible results. 

[balajij8]

Yes Karthik. It's a good model for numerous cases.