Cursor
Cursor supports custom OpenAI-compatible endpoints, so requests can be routed through Fenno.
Steps
- Open Settings → Models.
- Click Add model and enter an exact model ID from the console.
- Expand OpenAI API Key and paste your Fenno key.
- Set Override OpenAI Base URL to:
https://api.fenno.ai/v1- Save the key, enable Use API Key and the Base URL override, then run Verify if your version provides it.
- Start a new conversation, select the configured model explicitly and check its usage in the Fenno console.
Some Cursor features (for example inline completion) rely on specific upstream models. If a feature fails, switch that feature back to a default model; custom-key behavior depends on the client version, selected model and account policy. Not every feature is guaranteed to route through Fenno.
Troubleshooting
| Symptom | Fix |
|---|---|
| Verify fails | Base URL must include /v1 and reach api.fenno.ai |
| Model not found | Use the model id, not the display name |
| 429 | Lower concurrency or ask support for a higher quota |
Team policy and alternatives
Enterprise administrators can restrict personal API keys. If custom configuration is unavailable, check team policy before treating it as a Fenno authentication error. A custom key does not bypass the client’s own data processing; review its privacy documentation.
For independent provider configuration, run Claude Code or Codex in Cursor’s terminal without changing editors.
See the official Cursor settings updates and team model management .