Models and sign-in
Connect an API key, a server you run, or a subscription.
Adding a provider
In Settings → Model, choose Add provider, fill in each card and save it. There is one model chooser, grouped by provider. The model list is fetched from the provider, cached, and refreshed once per run or when you press refresh.
- Built in
- OpenAI, Claude, Gemini, OpenRouter, and any OpenAI-compatible endpoint, including one you run yourself
- Marketplace packages
- Other providers are added as provider packages from a marketplace.
Signing in to OpenAI
OpenAI offers three ways in: Sign in in browser, Use device code, and Use API-key chat. Sign-in URLs and tokens never appear on the settings page.
Using a subscription
- ChatGPT plan
- Sign in through an isolated local Codex runtime.
- Claude subscription
- Ave launches the official claude CLI you select. It uses an app-owned config folder and never reads OAuth tokens or reuses API keys. The CLI’s own tools are off; only Ave’s tools are offered, through its MCP bridge. This route has no model choice and no image attachments.
Subscription credentials never become API keys, and there is no silent fallback from one route to another.
Fallback and reasoning
With a fallback chain, the next model takes over a turn when one fails. For models that reason, Extended Thinking sets the reasoning level.