LLM Page Translator

LLM Page Translator turns any OpenAI-compatible API into a full-page web translator. Point it at OpenAI, a local model (llama.cpp, Ollama, LM Studio, vLLM), or any compatible endpoint, and read entire pages in the language you want — using the model of your choice. HOW IT WORKS - One click translates the whole page, preserving layout and structure. - Fast, parallel processing with a priority queue and an LRU cache (re-reading is instant). - Detects dynamic content and SPA navigation, so pages translate as you scroll. - Runs fully offline with a local model — nothing leaves your machine except your own API calls. KEY FEATURES - Any language, any direction (23 languages + Auto Detect) - Works with any OpenAI-compatible endpoint: OpenAI, OpenRouter, Groq, Mistral, local models, and more - Bring your own API key — it stays in your browser - Per-site settings: language pair, prompt, and CSS selectors for what to translate or skip - Controls: translate on demand, re-translate (fresh call), toggle original/translated per paragraph - Editable system prompts, with genre presets for Qidian web novels (e.g. Chinese → English or Vietnamese) - A DOM scanner suggests which parts of a page to translate or ignore PRIVACY All config and cache stay local (chrome.storage.local). No backend, no account, no tracking. The only requests the extension makes go to the endpoint you configure. If it speaks the OpenAI chat completions API, LLM Page Translator can use it.