AI Model Not Found - Model name is not valid: "auto"

[Wanderson (Wandersonleandro)]

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

When using Cursor AI, the application fails to process requests because the model selection defaults to “auto”, which is rejected as an invalid model name.

The error message displayed is:

“AI Model Not Found - Model name is not valid: ‘auto’”

This prevents AI Chat and Agent features from working. The issue occurs consistently and appears to be related to model resolution or account configuration.

Request ID: 3c64c108-bdac-4c88-b13b-6207306e92e7

Steps to Reproduce

1. Open Cursor IDE.
2. Open any project/workspace.
3. Open AI Chat or Agent.
4. Submit any prompt or request.
5. Cursor attempts to process the request.
6. The operation fails with the error:

“AI Model Not Found - Model name is not valid: ‘auto’”

The issue happens every time and prevents use of AI features.

Expected Behavior

Cursor should automatically resolve the selected model when “auto” is configured and successfully process AI requests.

Instead of failing with an “AI Model Not Found” error, the request should be routed to an available model and return a response normally.

Screenshots / Screen Recordings

Operating System

Linux

Version Information

Environment:

• OS: Linux
• Cursor Version: 3.13.25
• AI Features Affected: Chat and Agent

Additional Information

Environment:

• OS: Linux
• Cursor Version: 3.13.25
• AI Features Affected: Chat and Agent

Does this stop you from using Cursor

Yes - Cursor is unusable

[kevinn (Kevin Neilson)]

Hi @Wanderson Thank you for the forum post!

This is a known bug on our side, not something wrong with your account or team settings. After you hit a paid usage limit, the “Switch to auto” control can set the model to a value the backend does not accept, which produces the “Model name is not valid: auto” / “AI Model Not Found” error.

In the meantime, this workaround should unblock you:

1. Do not use the “Switch to auto” button in the error.
2. Open the model picker yourself (at the bottom of the chat).
3. Select Auto, then hover it and choose Edit, and set Optimize For to Cost (Auto Cost).
4. Alternatively, select plain Composer 2.5 (not the Fast variant).
5. Send your prompt again.

Choosing Auto Cost (or plain Composer 2.5) from the picker sets a valid fallback model. If you still hit an error after that, reply here with a fresh Request ID and a screenshot of the model picker, and I will take another look.

[Wanderson (Wandersonleandro)]

Thanks. It’s was resolved.