NEW: @MistralAI releases Mistral 3, a family of multimodal models, including three start-of-the-art dense models (3B, 8B, and 14B) and Mistral Large 3 (675B, 41B active). All Apache 2.0! 🤗 Surprisingly, the 3B is small enough to run 100% locally in your browser on WebGPU! 🤯

Link to collections: - Mistral Large 3 (675B, 41B active): https://t.co/Ei0H4GOxyv - Dense models (3B, 8B, and 14B): https://t.co/R9Nu2hUC8R Try out the demo yourself! 👇 https://t.co/vo2Llt05nX

Transformers v5's first release candidate is out 🔥 The biggest release of my life. It's been five years since the last major (v4). From 20 architectures to 400, 20k daily downloads to 3 million. The release is huge, w/ tokenization (no slow tokenizers!), modeling & processing.

Massive kudos to the @Supertone_ai team for this fantastic model release. I honestly can't believe something like this is finally possible! 🤩 🔗 Demo + source code:

Introducing Supertonic WebGPU: blazingly fast text-to-speech running 100% locally in your browser. ⚡️ Generate free audiobooks in seconds, no installation required. In this example, I created a 5-hour audiobook of The Great Gatsby in less than 3 minutes! 🤯 Try it yourself! 👇

SAM 3 can even run 100% locally in your browser on WebGPU, enabling real-time image segmentation! ⚡️ Original model: https://t.co/Vaiv6w2HAN ONNX model: https://t.co/ffy8aKC1Lk WebGPU demo: https://t.co/jjYfMOuC7Q

Meta Superintelligence Labs just released SAM 3: Segment Anything with Concepts. A unified model that detects, segments, and tracks objects in images or videos using concept prompts (text or images)! 🤯 Model weights (+ demo) now out on 🤗 Hugging Face!

I can't wait to see what the community builds with it! 🤗 Links: 📄 Blog post: https://t.co/qYI296GWLX 🚀 Demo (+ source code):

IBM just released Granite-4.0 Nano, their smallest LLMs ever (300M & 1B)! 😍 The models demonstrate remarkable instruction following and tool calling capabilities, and can even run locally in-browser! This means they can interact with websites and call browser APIs for you! 🤯

Just remember: nanochat isn't meant to be the most powerful LLM out there. Rather, it's designed as an educational resource to show you how to build a full-stack LLM from start to finish (< $1000). Enjoy! 🤗 Try it out yourself (+ source code):

BOOM! 💥 Today I added WebGPU support for @karpathy's nanochat models, meaning they can run 100% locally in your browser (no server)! The d32 version runs at over 50 tps on my M4 Max 🚀 Pretty wild that you can now deploy AI applications using just a single index.html file 😅

🔗 Demo + source code:

Introducing Granite Docling WebGPU 🐣 State-of-the-art document parsing 100% locally in your browser! 🤯 🔐 No data sent to a server (private & secure) 💰 Completely free... forever! 🔂 Docling ecosystem enables conversion to HTML, Markdown, JSON, and more! Try out the demo! 👇

🗂️ Model collection: https://t.co/QFBIJcG4DZ 🔗 WebGPU demo + source code:

IBM just released Granite 4.0, their latest series of small language models! These models excel at agentic workflows (tool calling), document analysis, RAG, and more. 🚀 The "Micro" (3.4B) model can even run 100% locally in your browser on WebGPU, powered by 🤗 Transformers.js!

Do you remember #Jarvis, the AI Assistant from Iron Man? I just built one entirely in #JavaScript as an #MCP-Client powered by #TransformersJS. It's a local-first assistant that runs completely in the browser on your machine. No API calls, no cloud dependency!

Transformers.js lets you run AI models offline in the browser with ONNX + WebGPU. Build a chatbot with Llama 3.2, optimize performance, and try local AI tasks like object detection & background removal.

Each document becomes a star, clustered together with other documents of similar meaning. Simply type a query, and fly through the galaxy to find the most relevant result. ⭐️ Learn more in our blog post: https://t.co/bj9g5OaORq 🔗 Link to demo:

NEW: Google releases EmbeddingGemma, a state-of-the-art multilingual embedding model perfect for on-device use cases! At only 308M params, the model can run 100% locally in your browser! 🤯 Explore your documents in an interactive 3D universe with our demo: "The Semantic Galaxy"