LocalLingo — Local AI Translator par thekyr

LocalLingo uses a local AI model — run with Ollama (or any OpenAI-compatible server) — to translate pages and text on your machine. It bundles no AI model and contacts no remote service — all translation requests go only to your local server at http://localhost / http://127.0.0.1.

HOW TO TEST
1. Install Ollama (https://ollama.com)
2. Pull a model, e.g. ollama pull qwen2.5:3b
3. Start Ollama with the browser origin allowed: OLLAMA_ORIGINS="*" ollama serve
(Without this, Ollama returns 403 and translations fail.)
4. Default server URL is http://localhost:11434 (configurable in options)
5. Set the Ollama model to use in the extension options
6. Full page translation: click the toolbar icon → pick a language → Translate Page
7. Selection translation: select text → right-click → Translate selection

Privacy
No analytics, no telemetry. Text to be translated is sent only to your local
model server and never to the author or any third party.