Skip to content

Use Cursor with the Gateway

Cursor is an AI-powered code editor built on VS Code. You can bring your own API key to route chat and agent requests through the Gateway, so they're processed by models in your chosen region with full spend controls.

Prerequisites

Connect Cursor to the Gateway

  1. Open Cursor Settings (⌘⇧J on macOS, Ctrl+Shift+J on Windows/Linux) and go to Models. This is not the VS Code-style Settings (⌘,).
  2. Under API Keys, enter your LLM API key in OpenAI API Key
  3. Turn on Override OpenAI Base URL and enter:
https://llm.[your-region].amazee.ai/v1

Replace [your-region] with your region code (e.g. ch103, us103).

  1. Click Verify to confirm the connection
  2. In the model list, click + Add Model and enter a model ID exactly as the Gateway lists it, e.g. claude-4-5-sonnet. Repeat for each model you want. See Available Models for the IDs in your region.
  3. Turn on the models you added, and pick one in the chat or agent panel

What goes through the Gateway

  • Chat and agent requests to the models you added go through the Gateway and are billed to your workspace.
  • Tab completion always uses Cursor's own models. It never goes through the Gateway.
  • Requests still pass through Cursor's servers on the way. Check Cursor's data policy if that matters for your compliance needs.

Troubleshooting

Verification fails
Make sure the base URL includes /v1 at the end and your API key is correct.
Model not available in picker
Cursor doesn't fetch the model list from the Gateway. Add each model ID with + Add Model, spelled exactly as Available Models shows it.
Cursor's built-in models stop working
Some Cursor versions send all OpenAI-style requests to the override URL. Turn Override OpenAI Base URL off when you want to use Cursor's own models.
Requests still going to OpenAI
Confirm the override base URL is saved. Restart Cursor if needed.