Skip to content

Set up local Ollama or a LAN translation server

The offline package starts a private Ollama runtime on http://127.0.0.1:11435 with the Qwen translation model. The runtime stops when the last app instance closes.

For your own server, select an LLM provider in Engines, enter its URL and model name, and choose Test → Apply.

Provider Example URL
Your local Ollama instance http://127.0.0.1:11434
Ollama on your LAN http://192.168.1.10:11434
OpenAI-compatible server on your LAN http://192.168.1.10:8000/v1

Start the server before testing the endpoint and make sure your PC can reach it. Ludivoxa uses the endpoint you configure while translating. Update checks use GitHub separately and can be turned off. Shared local and LAN servers run independently of the app.