AnyLLMTranslate

Bilingual web page translation powered by any OpenAI-compatible LLM

AnyLLMTranslate reads the web bilingually. It translates pages, video subtitles, selected text, input fields, and PDFs into the language you choose — and it does it with your own LLM API key, not ours. BRING YOUR OWN KEY There is no AnyLLMTranslate server. You point the extension at any OpenAI-compatible endpoint — a cloud provider or a local runtime such as Ollama or LM Studio — and your text goes straight there. Your API key is encrypted in your browser and never leaves it except to the provider you chose. No account, no sign-up, no proxy. WHAT IT DOES • Bilingual pages — show the original and the translation together, above, below, beside, or translation-only. Visible content is translated first, with batching, streaming, caching, and prefetch, and single-page apps keep working as you navigate. • Video subtitles — translate captions progressively and show them in the player or in a resilient overlay, with reading-speed timing, line wrapping, and per-site profiles. • Selection and hover — select a word for a dictionary-style definition or a sentence for a focused translation; hover a paragraph to translate it after a delay you set. • Inline input — translate what you type in inputs, textareas, and contenteditable fields with a key gesture or a keyboard shortcut. • PDF — open PDFs in a bundled reader, and optionally translate scientific PDFs through a local Docker bridge you run yourself, preserving layout, math, and figures. • Your terms — named glossaries, per-site rules and excludes, custom prompts, themes, and keyboard shortcuts. PRIVACY • No telemetry, no analytics, no advertising, no affiliate links. • Your API key is stored encrypted in your browser and is only ever sent to the provider you configured. • The first-run setup shows you exactly which data leaves your browser and where it goes, and asks you to accept it before anything is translated. • Local usage statistics stay on your device and can be exported or erased at any time. WHAT YOU NEED An API key and an endpoint from an OpenAI-compatible provider, or a local runtime such as Ollama or LM Studio. PDF translation additionally requires the optional local bridge, which is off by default. Open source under the MIT license.