Enabling OpenAI API Key makes OpenAI thing it's using deepseek-r1 model

Bug Reports
1

:lady_beetle: Provide a clear description of the bug
If I enter an OpenAI API key and have deepseek-r1 enabled as a model, OpenAI will try to use deepseek. Also, I cannot use deepseek-r1 while my OpenAI API key is enabled.

Output:
curl https://api.openai.com/v1/chat/completions -H “Content-Type: application/json” -H “Authorization: Bearer xxxxxxx” -d ‘{
“messages”: [
{
“role”: “system”,
“content”: “You are a test assistant.”
},
{
“role”: “user”,
“content”: “Testing. Just say hi and nothing else.”
}
],
“model”: “deepseek-r1” <—Why is it using deepseek-r1??
}’

Also, you cannot use the deepseek-r1 model anymore. It will give an error. If you disable your OpenAI API Key, then you can use deepseek-r1 with no issues.

Because of this, it’s not possible to use deepseek-r1 and have an OpenAI API Key enabled

:arrows_counterclockwise: Explain how to reproduce the bug (if known)

  1. Have deepseek-r1 model enabled
  2. Perform a test query in Chat and see it’s working fine.
  3. Enable OpenAI API key and enter a key and press Verify
  4. You get an error saying the model is not valid for your API key
  5. Notice how at the bottom of the error message it’s using deepseek-r1 as the model
  6. Enable the gpt-3.5-turbo model
  7. Enable and verify your OpenAI API key
  8. It will succeed now
  9. Try to run a query in Chat with deepseek-r1
  10. It will say “The model deepseek-r1 does not work with your current plan”

:computer: Tell us your operating system and your Cursor version (e.g., Windows, 0.x.x).
macOS 15.3.1 (24D70)
Version: 0.45.11
VSCode Version: 1.96.2
Commit: f5f18731406b73244e0558ee7716d77c8096d150
Date: 2025-02-07T09:43:58.555Z
Electron: 32.2.6
Chromium: 128.0.6613.186
Node.js: 20.18.1
V8: 12.8.374.38-electron.0
OS: Darwin arm64 24.3.0

2

Hey, this is an error verifying your key, it sees an unfamiliar model, which is why it outputs it. Try disabling all models not related to OpenAI and try again. As for the inability to use the DeepSeek R1 model while your key is active, it’s because all requests go through a single thread. You need to disable your API key every time you need a model from your subscription.