TokenFleet
Tutorials

Cursor

Configure Cursor with an OpenAI-compatible gateway connection.

Before you start

Cursor usually connects to custom model services through an OpenAI-compatible API. What you need most is:

  1. Your API gateway domain
  2. An OpenAI-compatible base URL
  3. A valid token
  4. A backend-exposed model name

Your docs domain and API address can be different.

Install Cursor

Cursor is typically installed as a desktop client, then configured from the in-app settings.

Configure the provider

In Cursor's model or provider settings, fill in:

  • API key: sk-your-token
  • Base URL: https://api.example.com/v1
  • Model: the model name exposed by your backend

If your Cursor version supports a custom OpenAI provider, this is usually enough.

Validation

Test the endpoint first from a terminal:

curl https://api.example.com/v1/models \
  -H "Authorization: Bearer sk-your-token"

Then return to Cursor and save the same values.

Troubleshooting

No models appear

Your /v1/models response may be empty, or the token may not have access.

The model appears but requests fail

Confirm that the model actually supports the feature you are trying to use.

Root URL fails

For OpenAI-compatible mode, the base URL usually needs the /v1 suffix.