UnioAPI Documentation
Client Setup

Cursor API Setup

Configure a UnioAPI API key and custom OpenAI Base URL in Cursor. Follow the setup steps to connect available models and verify API access.

Setup steps

  1. Open Cursor Settings → Models → API Keys.

  2. Paste your UnioAPI key into OpenAI API Key and enable it.

  3. Turn on Override OpenAI Base URL and enter:

    https://api.unioapi.com/v1

  4. Back in the Models list, click Add model and enter an available model identifier (e.g. gpt-5.6-sol; full list: GET /v1/models).

  5. Pick that model in chat and send a message — a normal reply means you are connected.

Notes

  • The Override switch is global: while it is on, every OpenAI-family request (including models picked from the built-in list) goes to this endpoint. Turn it off temporarily to use Cursor-managed models.
  • Cursor sends Responses API payloads in some scenarios; UnioAPI serves both POST /v1/chat/completions and POST /v1/responses, so both are handled.

Troubleshooting

SymptomCheck
Network error / TLS disconnectTry Cursor Settings → Network → HTTP Compatibility Mode, switch to HTTP/1.1.
401 authentication failedConfirm the API key is enabled and not revoked.
Responses come from the wrong backendCheck the Override switch — it decides whether requests go to UnioAPI or Cursor's backend.

Codex API Setup

From fresh install to UnioAPI via config.toml, over the Responses API.

TRAE API Setup

Add a custom model — note the base address must not include /v1.

On this page

Setup stepsNotesTroubleshooting