Cursor
Use your PromptsForLess key in Cursor chat by overriding the OpenAI base URL and adding a custom model ID.
Cursor lets you use your own API key for chat models. Tab completion always uses Cursor's own models. You need a Cursor version that shows the Override OpenAI Base URL setting.
Cursor's API key documentation covers which features use your key.
Add your key
- Open Cursor Settings → Models.
- Find the OpenAI API Key controls.
- Enter your PromptsForLess key from the dashboard.
- Turn on Override OpenAI Base URL and set it to
https://api.promptsforless.com/v1. - Add a custom model with the exact ID from the catalog, for example
deepseek-v4.1-flash.
Put your PromptsForLess key in the OpenAI field only. Other providers' key fields send requests to those providers, not to PromptsForLess.
Check chat first
Select the model in a new chat and send a short prompt. Once it replies, try agent actions or attach files. For agent work, pick a model with tool calling.
To rule out editor settings, send a direct request with the same key.
Know what you pay for
Your PromptsForLess balance pays for model requests. Your Cursor plan is billed by Cursor. Cursor sends API-key requests through its own servers to build prompts, so read its privacy documentation before setting up a team workspace. On our side, we keep usage and billing metadata only, never prompts or responses.
Troubleshooting
| Symptom | Fix |
|---|---|
| No base URL override | Use a tool with a custom provider setting, such as OpenCode. |
| Key verification fails | Check the key and base URL, then send a direct request. |
| Model missing from the picker | Add it as a custom model with the exact ID. |
| Tab still uses Cursor | Expected: your key covers chat models, not Tab. |
Compare model prices in the catalog.