# TranslateGemma 4B WebGPU

> Private, in-browser machine translation running Google's TranslateGemma model via WebGPU

TranslateGemma 4B WebGPU is a browser-based translation demo powered by Google's TranslateGemma model and Transformers.js. All processing runs client-side using WebGPU and ONNX Runtime Web, meaning translations are completely private and work offline once the model finishes downloading. It supports translation across 56 languages without sending text to an external server.

Website: https://huggingface.co/spaces/webml-community/TranslateGemma-WebGPU

## What it is for

- Translate confidential or sensitive text without sending data to third-party servers
- Perform offline machine translation in the browser without an active internet connection
- Test on-device WebGPU inference capabilities with large language models in browser environments

## Features

- Client-side inference using Transformers.js and ONNX Runtime Web
- Hardware acceleration via WebGPU
- Privacy-focused architecture with zero server-side data transfer
- Offline functionality once initial model assets are cached
- Support for translation across 56 languages

## Pricing

Free to use directly in the browser

## Details

- Type: Web app
- Category: [AI](https://xiry.app/topics/ai.md)
- Open source: yes
- Platforms: Web
- For: Developers, researchers, and privacy-conscious users needing local text translation
- Alternative to: Google Translate, DeepL
- Tags: translation, webgpu, transformers js, onnx, local ai, nlp

## Why it was saved

- [@nicodotdev, 24 February 2026](https://x.com/nicodotdev/status/2026367494886199722): Free, private, offline-capable translation across ~55 languages with no server and no data sent anywhere; model is a 2.9GB download and requires a WebGPU-capable browser with sufficient VRAM.

## Related

- [TranslateGemma (google/translategemma-4b-it)](https://xiry.app/resources/translategemma-google-translategemma-4b-it): Open translation model by Google DeepMind lightweight enough for on-device use. (AI model, Open source)
- [HeyGen](https://xiry.app/resources/heygen): AI avatar video generation and translation platform driven by text prompts (Web app, Freemium)
- [Pinokio](https://xiry.app/resources/pinokio): Browser that installs and runs open-source AI apps with one click (Desktop app, Free)
- [Local Deep Researcher](https://xiry.app/resources/local-deep-researcher-2395): Fully local web research and report writing assistant using Ollama or LMStudio (GitHub repo, Open source)
- [AgenticSeek](https://xiry.app/resources/agenticseek): Open-source, fully local autonomous AI agent that browses and writes code (GitHub repo, Open source)
- [AnythingLLM](https://xiry.app/resources/anythingllm): Private, on-device AI assistant to chat with documents and local models. (Desktop app, Open source)

Updated 10 October 2026. Page: https://xiry.app/resources/translategemma-4b-webgpu

Make your own library: Xiry takes your saved reels, tweets, Reddit bookmarks and Watch Later videos, makes them searchable and pulls the useful stuff out, like this library. It's invite-only for now; join the waitlist at https://xiry.app/#waitlist.