Kokoro WebGPU: Real-time text-to-speech running 100% locally in your browser.

Introducing Kokoro Web: ML-powered speech synthesis directly in your browser. Now with streaming & WebGPU acceleration.

Janus Pro 1B running 100% locally in-browser on WebGPU, powered by Transformers.js

SmolVLM 256M: The world's smallest multimodal model, running 100% locally in-browser on WebGPU.

DeepSeek-R1-Distill-Qwen-1.5B running 100% locally in-browser on WebGPU. Reportedly outperforms GPT-4o and Claude-3.5-Sonnet on math benchmarks (28.9% on AIME and 83.9% on MATH).

Introducing Kokoro.js: a new JavaScript library for running Kokoro TTS (82M) locally in the browser w/ WASM.

WebGPU-accelerated reasoning LLMs running 100% locally in-browser w/ Transformers.js

Vision Transformer Explorer: interactively explore the self-attention maps produced by ViTs

Moonshine Web: Real-time in-browser speech recognition that's faster and more accurate than Whisper

TTS WebGPU: The first ever text-to-speech web app built with WebGPU acceleration (powered by OuteTTS and Transformers.js)

Janus, a new multimodal understanding and generation model from Deepseek, running 100% locally in the browser on WebGPU with Transformers.js!

Transformers.js v3 is finally out: WebGPU Support, New Models & Tasks, New Quantizations, Deno & Bun Compatibility, and More…

OpenAI's new Whisper Turbo model running 100% locally in your browser with Transformers.js

Running Llama 3.2 100% locally in the browser on WebGPU w/ Transformers.js

LLAMA3.2

Phi-3.5-mini running in-browser at ~90 tokens/second on WebGPU w/ Transformers.js and ONNX Runtime Web.

Whisper Diarization Web: In-browser multilingual speech recognition with word-level timestamps and speaker segmentation

Whisper Timestamped: Multilingual speech recognition w/ word-level timestamps, running locally in your browser using Transformers.js

How to run Gemini Nano locally in your browser with Google Chrome's new Prompt API.

Florence-2 WebGPU: Powerful vision foundation model running locally in your browser w/ Transformers.js

Whisper WebGPU: Blazingly-fast ML-powered speech recognition directly in your browser

WebGPU-accelerated real-time in-browser speech recognition w/ Transformers.js

Firefox will use on-device ML to power translation and image alt text generation

Moondream WebGPU: your favorite tiny vision language model now runs 100% locally in the browser

Phi-3 WebGPU: a private and powerful AI chatbot that runs 100% locally in your browser