Cursor can use custom model providers in current product builds, but Cursor’s public docs were not specific enough to verify every field label from this environment. Use the steps below as an in-product setup guide, and confirm labels against your current Cursor build. Official reference used:Documentation Index
Fetch the complete documentation index at: https://docs.routing.run/llms.txt
Use this file to discover all available pages before exploring further.
Setup
Get your API key
Create an API key at app.routing.run. Copy the key — it starts with
rk_.Open Cursor settings
Open Cursor and press
Cmd+, (macOS) or Ctrl+, (Windows/Linux). Navigate to Models.Add routing.run as a custom provider
In the custom provider or model provider settings, create a saved provider entry named
routing.run if your Cursor version offers a label field. Paste your full secret from app.routing.run → API Keys into the API key field, then set the base URL to https://api.routing.run/v1 with no trailing slash.Manual configuration
If you want to fill Cursor settings manually, use these exact values:| Field | Value |
|---|---|
| Saved provider name | routing.run if the UI offers a label or provider name field |
| OpenAI API Key | rk_REPLACE_ME |
| OpenAI Base URL | https://api.routing.run/v1 |
| Model | route/glm-5.1 |
route/ prefix on model ids. If Cursor lets you save multiple custom providers, name this one routing.run so it is easy to switch back to it later.
Cursor’s public docs did not expose a stable field-by-field reference for this flow in this environment. If your current build uses different labels, match the same underlying values: API key, base URL, and
route/… model id.Recommended models
Published model IDs are listed on the models page. Exact access depends on your plan tier and the dashboard; common picks:| Model | Use case |
|---|---|
route/glm-5.1-precision | Strongest reasoning-first choice |
route/qwen3.6-plus | Best default starting point |
route/minimax-m2.7-highspeed | Long-context sessions |
route/kimi-k2.5 | Agentic and tool-heavy flows |
route/glm-5.1 | Faster GLM alternative |
route/qwen3.5-plus | Qwen3.5 flagship |
Troubleshooting
Connection refused or timeout
Connection refused or timeout
Make sure the base URL is exactly
https://api.routing.run/v1 — no trailing slash. Check that your API key is correct.Model not found error
Model not found error
The model name must include the
route/ prefix. For example, use route/glm-5.1, not glm-5.1.Model not allowed error
Model not allowed error
Some models are only available on paid plans. Check available models for your tier.