Description
Want translation for your browser extension or reading foreign material without sending content to someone's server every time, or hitting paid quotas? MTranServer runs the whole thing on your own machine—an ultra-low-footprint, very fast offline translation model server averaging about 50ms per request, and it needs no GPU. An ordinary PC can host an unlimited free translation service.
It focuses on offline, fast, self-hostable: the desktop build launches in one click with a tray menu and a simple UI, and you can also deploy it as a server with one npx line or via Docker, exposing an API for Immersive Translate and similar extensions. Quality won't match online large models, but it wins on local, free, private and fast — plenty for everyday web and document translation. For a translation backend entirely your own, pick it.
Offline translation: runs the model locally — no internet, no uploads, so even sensitive text never leaves your machine.
Very fast: about 50ms per request on average, no GPU needed, light enough for an ordinary PC or even a soft router.
One-click desktop: download and run on Windows/Mac/Linux, managed from a tray menu with a built-in simple UI and debug docs.
Immersive Translate ready: exposes an API to plug into Immersive Translate and similar extensions, sending whole-page and document translation to your own server.
Private deployment: start with one npx line or deploy via Docker to run your own unlimited free translation service.
Many languages: translates among the world's major languages.
It focuses on offline, fast, self-hostable: the desktop build launches in one click with a tray menu and a simple UI, and you can also deploy it as a server with one npx line or via Docker, exposing an API for Immersive Translate and similar extensions. Quality won't match online large models, but it wins on local, free, private and fast — plenty for everyday web and document translation. For a translation backend entirely your own, pick it.
Features
Offline translation: runs the model locally — no internet, no uploads, so even sensitive text never leaves your machine.
Very fast: about 50ms per request on average, no GPU needed, light enough for an ordinary PC or even a soft router.
One-click desktop: download and run on Windows/Mac/Linux, managed from a tray menu with a built-in simple UI and debug docs.
Immersive Translate ready: exposes an API to plug into Immersive Translate and similar extensions, sending whole-page and document translation to your own server.
Private deployment: start with one npx line or deploy via Docker to run your own unlimited free translation service.
Many languages: translates among the world's major languages.
