Non-Auto Models Not Working

Support Bug Reports
, ,
1

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

Whenever I try to send a message to an agent that’s not set to “Auto”, I get errors as if I don’t have access to that model (I just upgraded to the $60 plan, not a limit issue). For example, I tried to send a simple request to Composer 2, I get a message saying the model doesn’t support custom API keys:

Request ID: 8ea6d605-2037-4cdd-860d-4cc5fafdcddf
{“error”:“ERROR_BAD_REQUEST”,“details”:{“title”:“Bad Request”,“detail”:“This model does not support custom API keys.”,“isRetryable”:false,“additionalInfo”:{},“buttons”:,“planChoices”:,“analyticsMetadata”:{“actionRequired”:“config”}},“isExpected”:true}
Bad Request This model does not support custom API keys.
l5t: Bad Request This model does not support custom API keys.
at Pf_ (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29432:23999)
at Rf_ (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29432:23144)
at Uf_ (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29433:6490)
at q0h.run (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29433:11307)
at async cQi.runAgentLoop (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:41748:12418)
at async Vnm.streamFromAgentBackend (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:41816:12344)
at async Vnm.getAgentStreamResponse (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:41816:19567)
at async lCt.submitChatMaybeAbortCurrent (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29548:16906)
at async Object.Va [as onSubmit] (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:40804:4230)
at async vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:40803:100980

When I try GPT5.5, it keeps saying “reconnecting” then I get this:

Request ID: 24cc5131-f84a-49ce-a876-b92729084805
{“error”:“ERROR_OPENAI”,“details”:{“title”:“Unable to reach the model provider”,“detail”:“We encountered an issue when using your API key: You exceeded your current quota, please check your plan and billing details. For more information on this error, read the docs: https://platform.openai.com/docs/guides/error-codes/api-errors.\n\nAPI Error:\n\n\n{\"type\":\"error\",\"error\":{\"type\":\"insufficient_quota\",\"code\":\"insufficient_quota\",\"message\":\"You exceeded your current quota, please check your plan and billing details. For more information on this error, read the docs: https://platform.openai.com/docs/guides/error-codes/api-errors.\",\"param\":null},\"sequence_number\":2}\n”,“additionalInfo”:{},“buttons”:,“planChoices”:},“isExpected”:true}
[invalid_argument] Error
mIe: [invalid_argument] Error
at Pf_ (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29432:24345)
at Rf_ (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29432:23144)
at Uf_ (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29433:6490)
at q0h.run (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29433:11307)
at async cQi.runAgentLoop (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:41748:12418)
at async Vnm.streamFromAgentBackend (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:41816:12344)
at async Vnm.getAgentStreamResponse (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:41816:19567)
at async lCt.submitChatMaybeAbortCurrent (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29548:16906)
at async Object.Va [as onSubmit] (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:40804:4230)
at async vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:40803:100980

Steps to Reproduce

  1. Add a message to the agent.
  2. Switch to any model that isn’t “Auto”.
  3. Error message

What I’ve Tried That Didn’t Work

  1. Reinstall Cursor
  2. Clear cache
  3. Update Cursor

Only the older versions (not anything in the 3 range) works but I lose my chats from 3 if I’m using 2.

Expected Behavior

The agent should complete the requested task.

Operating System

MacOS

Version Information

Version: 3.2.11
VSCode Version: 1.105.1
Commit: e9ee1339915a927dfb2df4a836dd9c8337e17cc0
Date: 2026-04-24T14:36:47.933Z
Layout: editor
Build Type: Stable
Release Track: Default
Electron: 39.8.1
Chromium: 142.0.7444.265
Node.js: 22.22.1
V8: 14.2.231.22-electron.0
OS: Darwin arm64 24.6.0

For AI issues: which model did you use?

Composer 2, GPT 5.5, Codex 5.3

Does this stop you from using Cursor

No - Cursor works, but with this issue

5

Hey, this isn’t a bug. You have a custom OpenAI API key enabled in Cursor Settings → Models, so non-Auto models are routed through it. That’s why you’re seeing both errors:

  • Composer 2: this is a Cursor proprietary model, so it won’t work with external API keys. That’s why you get This model does not support custom API keys.
  • GPT 5.5: the request went to OpenAI using your key, and OpenAI returned insufficient_quota, meaning your OpenAI account is out of quota or billing isn’t set up.

What to do: open Cursor Settings → Models and turn off or remove your OpenAI API key. After that, Auto and all non-Auto models will use your Cursor Pro+ limits. If you really want BYOK, add funds on OpenAI, but Composer 2 still won’t work with your key. That’s by design.

Let me know if that fixed it.