AI Model Not Found Model name is not valid: "deepseek-v4-pro"

Support Bug Reports
1

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

When I use the deepseek-v4-pro model, after the response is completed, I get the error “Unable to resume after background task completion: AI model not found. Invalid model name: ‘deepseek-v4-pro’”. Why does this error not occur when using deepseek-v4-flash? Furthermore, when this error occurs, the OpenAI API Key switch will automatically turn off. If I want to continue using it, I must turn the switch back on in the models settings.

Steps to Reproduce

If you ask a complex question that’s broken down into multiple background subtasks, this error message will appear when the background tasks return. And judging from the error message, it seems the error occurred during the background task’s return process.

Operating System

MacOS

Version Information

Version: 3.2.21 (Universal)
VSCode Version: 1.105.1
Commit: 806df57ed3b6f1ee0175140d38039a38574ec720
Date: 2026-05-03T01:46:14.413Z
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 x64 24.6.0

For AI issues: which model did you use?

model name: deepseek-v4-pro

For AI issues: add Request ID with privacy disabled

Request ID:ae4e1750-af9a-4d39-82b1-3bb2e48924c4

Additional Information

Request ID: ae4e1750-af9a-4d39-82b1-3bb2e48924c4
m5t: AI Model Not Found Model name is not valid: “deepseek-v4-pro”
at Gv_ (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29432:23999)
at Wv_ (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29432:23144)
at Xv_ (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29433:6490)
at UCh.run (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:29433:11307)
at async fQi.runAgentLoop (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:41749:12418)
at async nrm._runBackgroundTaskCompletionAction (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:41817:15063)
at async nrm._maybeDispatchBackgroundCompletions (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.desktop.main.js:41817:17365)

Does this stop you from using Cursor

No - Cursor works, but with this issue

3

I’ve been using deepseek-v4 since its release, but this issue hasn’t been fixed to this day, even though I’m a cursor Pro subscriber.

6

Hey, thanks for the report with the stack trace and Request ID, that really helped.

Looks like you’re on 3.2.21 from 2026-05-03, and the fix for this case BYOK model losing context when resuming after a background subtask landed a bit later. Please update Cursor to the latest stable version and check if it still happens.

If it still happens after updating, post a new Request ID here with Privacy Mode turned off and we’ll take a look to see if it’s an edge case.

About the OpenAI API Key switch turning itself off after an error, I see that and I’ll flag it separately. If it still reproduces on the new version, let me know in this thread.

7

It’s possible that the version I was using just happened to have this bug. After updating, the “AI model not found” error no longer occurred.

8

Great, glad the error went away after the update. That means the fix for this case made it into the new version.

One thing: you mentioned before that after this error, the OpenAI API Key switch in the model settings would turn itself off. Is that still happening on the latest version? If yes, post here, and ideally include a new Request ID with Privacy Mode turned off, and we’ll look into it separately.