Using Letronna
Gateways
Where Letronna gets its models: your Letronna account, OpenAI, OpenRouter, Ollama or any OpenAI-compatible server.
A gateway is where the models come from. Set up as many as you like; the model menu lists them all, and each message goes to the gateway of the model you picked. See them in Settings > Accounts in the app, or with /gateways in the terminal.
| Gateway | Address | Needs |
|---|---|---|
| Letronna | https://api.letronna.ai/v1 | signing in; paid from your credit |
| OpenAI | https://api.openai.com/v1 | an OpenAI API key |
| OpenRouter | https://openrouter.ai/api/v1 | an OpenRouter key; hundreds of models |
| Ollama | http://localhost:11434/v1 | Ollama running on this computer; works offline |
| Custom | any OpenAI-compatible server | its address, and a key if it asks for one |
Keys are kept in keys.json in your data folder, on your disk only, and never reach the model. Your own keys cost nothing on Letronna's side: you pay the provider directly.
You can also connect Anthropic with an Anthropic API key, or sign in with a Claude subscription to use the models your plan includes, with no key to paste. When your model is overloaded or out of credit, Letronna can carry on with a fallback model you name.