Cursor

Cursor lets you bring your own OpenAI key and override the API address, which is how you use this site's models in it. Everything happens in Cursor's settings screen.

• Protocol: OpenAI-compatible
• Base URL: https://tokenfunc.com/v1
• Models: GPT, DeepSeek, Gemini and others, for example gpt-5

1. Open the model settings

   Open Cursor, go to Cursor Settings → Models, scroll to the bottom and expand API Keys.

   Screenshot pendingThe bottom of the Models page in Cursor Settings with the API Keys section expandedtools/cursor-api-keys.png

2. Enter the key and address

   1. Paste your key into OpenAI API Key: sk-xxxxxxxxxxxxxxxx
   2. Turn on Use OpenAI API Key.
   3. Turn on Override OpenAI Base URL and enter: https://tokenfunc.com/v1

   Screenshot pendingOpenAI API Key filled in, both switches on, and the base URL enteredtools/cursor-override.png

3. Add the model

   In the model list on the same page click + Add Model, type a model name from this site such as gpt-5, press Enter and make sure it is checked.

   The name must match the Model Square exactly.

   Screenshot pendingThe Add Model field in the model list with a model name from this sitetools/cursor-add-model.png

4. Verify

   Restart Cursor, start a new chat, choose the model you just added and send a message. A reply means it works.

   The Verify button in settings only checks connectivity, not whether the model name exists, so trust the actual chat.

Troubleshooting