Integrations
Cursor
Run Cursor's chat models on any model in the MiniRouter catalog.
Cursor is an AI code editor. Its OpenAI key and base URL override send chat requests to MiniRouter's Chat Completions endpoint, so chat runs on any model in the catalog, with one balance and per-key spend limits.
1. Create a key
Create a key for Cursor alone.
2. Check the key
One request outside Cursor proves the key and balance work.
curl https://api.minirouter.sh/v1/chat/completions \
-H "Authorization: Bearer mr-live-YOUR-KEY-HERE" \
-H "Content-Type: application/json" \
-d '{"model":"zai/glm-5.3-flash","messages":[{"role":"user","content":"ping"}]}'3. Enter the settings
Open Cursor Settings → Models → OpenAI API Key → Override OpenAI Base URL.
- OpenAI API Key
mr-live-… (your key)- Override OpenAI Base URL
https://api.minirouter.sh/v1This override applies globally to OpenAI-shaped model traffic in Cursor.- Model ID
zai/glm-5.3-flashCompatibility depends on Cursor's global OpenAI override.
4. Send a prompt
Pick zai/glm-5.3-flash in the chat model menu and send a prompt. Then open Activity. The request lists its model, tokens and exact cost.
What your key serves
- Chat modelsRun on MiniRouter with your key.
- Tab completionStays on Cursor's built-in models.
Add more models
Add any catalog ID as another Model ID. Copy IDs from Models.
Teams and privacy
- Teams and Enterprise
- Cursor still charges its Token Rate, $0.25 per million tokens.
- Individual plans
- Own-key requests do not use included usage.
- Zero Data Retention
- Does not apply. MiniRouter's privacy terms do.
Giving a team access? See Teams.
Keep spend in check
Restrict models or filter content with Guardrails.
Recommended modelsBrowse every model in the catalog.
zai/glm-5.3-flash- Best current balance of coding ability, agentic performance, context and cost.
openai/gpt-5.6-sol- Higher-intelligence option when task quality matters more than spend.
spacexai/grok-4.6- Strong agentic alternative with a 500K-token window.
google/gemini-3.8-flash- Multimodal coding alternative with a 1M-token window.
Troubleshooting
Built-in Cursor models stop working while the override is enabled
Cursor's OpenAI base-URL override is global. Turn it off when switching back to models served by Cursor rather than MiniRouter.
The override works, but requests still pass through Cursor
Cursor builds the final request on its own backend before forwarding it to the overridden endpoint. This is not a direct client-to-MiniRouter connection.
Override OpenAI Base URL is missing
Newer Cursor builds do not show it on every plan. Without it, Cursor cannot reach MiniRouter.
Tab completion ignores MiniRouter
Expected. Tab always uses Cursor's own models; your key serves chat models only.
404, HTML, or a parse error on every request
Almost always the base URL. Use https://api.minirouter.sh/v1 — the bare host without /v1 also works, but never append /chat/completions to the base URL field; the client adds that path itself.
Error: 400 invalid_request.
401 on every request
Wrong or rotated key. Keys start with mr-live-; after a rotation the old key keeps working for 24 hours, then dies.
Error: 401 invalid_api_key.
Every error code, with its fix: Errors.