Pre-releaseGantry has not shipped yet. These pages describe the designed behaviour and will be revised with the first release.
FAQ
Is there an account?
Section titled “Is there an account?”No. You need an API key from a model provider, and that is all.
Does anything get sent to Gantry?
Section titled “Does anything get sent to Gantry?”No. There is no Gantry server. Requests go from your machine to the provider or service you configured. The website makes two optional requests to GitHub, for the star count and the latest release.
Which models can I use?
Section titled “Which models can I use?”Anthropic, OpenAI, Google, xAI and OpenRouter models with your keys, local models through Ollama, and any OpenAI-compatible endpoint.
Can I use it offline?
Section titled “Can I use it offline?”The app runs offline; chats need a model, so with Ollama you can work with no network at all.
Is it open source?
Section titled “Is it open source?”Source-available under the Functional Source License, which converts each release to Apache 2.0 after two years. You can read, build, modify and share it.
How do I request a connector?
Section titled “How do I request a connector?”Open a connector request on GitHub. The app will have the same link in its Connectors window.