Letronna verwendenLocal models
Ollama
Using a model you run with Ollama: install, pull a model, pick it in Letronna.
Diese Seite ist noch nicht übersetzt und wird deshalb auf Englisch gezeigt. Du kannst sie auf GitHub übersetzen. GitHub ↗
Ollama runs open models on your computer. Letronna looks for it at http://localhost:11434/v1 and lists its models with no setup: no key, no address to type.
Set it up
Install Ollama from ollama.com and start it.
Download a model that can call tools, for example:
ollama pull qwen3:8bPick the model from the model menu. Its models are listed under Ollama, beside your other gateways.
On first launch, choose On this computer on the setup page; its Local model row says "Detected" with the number of models once Ollama answers, or "Not found" until it does. In the terminal, the same choice is in letronna setup, and /gateways shows Ollama among the rest.
When it does not answer
Ollama is listed only while it is running. If its models are missing from the menu, start Ollama and open the menu again. Then:
| It says | What to do |
|---|---|
| model not found, try pulling it first | ollama pull the model, with the name exactly as the menu shows it |
| the runner process has terminated | the model did not fit in memory, or crashed; try a smaller one |
| the conversation is too long | see the context window |
letronna doctor --offline checks the rest of the setup without the network; see Troubleshooting.