I’m experiencing several issues with accessing different AI models in Cursor:
-
I have 500 unused fast requests in my quota, but all my queries appear to be going through my own API keys (marked as “User API Key” in the usage history).
-
When trying to use deepseek-r1, I receive this error:
Request failed with status code 404: {
"error": {
"message": "The model `deepseek-r1` does not exist or you do not have access to it.",
"type": "invalid_request_error",
"param": null,
"code": "model_not_found"
}
}
-
When attempting to use gemini-2.0-flash, I get “Unauthorized User API Key” errors, despite my API being verified in the panel.
-
OpenAI and Claude models work via API, but Claude 3.7 sometimes has availability issues, forcing me to temporarily switch to OpenAI’s o3 model (which I personally find less effective).
I’m logged in with my correct email and can see my dashboard. Has anyone else encountered similar problems or know how to resolve these issues?