Joffrey Thomas

Introducing Voxtral WebGPU: Real-time speech transcription entirely in your browser. This demo runs Voxtral-Mini-4B, a powerful streaming ASR model from @MistralAI, locally on WebGPU. The model supports 13 languages and is capable of <500 ms latency. Fully private. Zero cost.

@Vassivasss @haaspierre_ what a commit ! Glad you participated !

What if AI could see the world the way we do? That’s the idea we bet our weekend on at the Mistral Worldwide Hackathon. With @haaspierre_ and Arman Artola-Zanganeh, we built 𝗣𝗼𝗿𝘁:𝗪𝗼𝗿𝗹𝗱🌍, an open-source framework that lets anyone connect their Meta glasses to any AI system. Let me take you back to saturday morning. So before knowing it could work we needed the hardware. So I ran to Rue de Rivoli and bought €500 Meta glasses on the spot. If that’s not commitment, I don’t know what is (a true bet). We then built non-stop for 36 hours to make it usable. End-to-end. The glasses stream what you see → the AI makes sense of it → it answers back through the glasses’ speaker. And suddenly when we understood that it was going to work, the question changed. It was no longer “𝗜𝘀 𝘁𝗵𝗶𝘀 𝗱𝗼𝗮𝗯𝗹𝗲?” It became “𝗪𝗵𝗮𝘁 𝗰𝗮𝗻 𝗽𝗲𝗼𝗽𝗹𝗲 𝗯𝘂𝗶𝗹𝗱 𝘄𝗶𝘁𝗵 𝘁𝗵𝗶𝘀?” - A plumber getting live assistance while repairing something. - A technician repairing industrial machinery. - A traveler exploring a new country. - A visually impaired person navigating space. At first, we were looking for the “right” use case. Then we realized something more interesting. If AI can share your perspective, continuously, the use cases are not ours to decide. That’s why 𝗣𝗼𝗿𝘁:𝗪𝗼𝗿𝗹𝗱🌍 is fully open source. If you want to connect your Meta glasses, plug in your own models, customize with your own prompts, your own MCP, your Openclaw… you can. Link to the open source repo (you can contribute and give it a little star ❤️): lnkd.in/es6YPSXe Link to the demo video: lnkd.in/ePTeJ486 Huge thanks to the organizing team of the hackathon, it was truly great. @Jthmas404

@nukonuko 素晴らしいプロジェクトです。ローカルのVoxtral MLX製品をリリースされることを楽しみにしています。

Mistral Worldwide Hackathon、有意義な時間でした！ #MistralAI 私のチームではハンズフリーの音声ガイド付きお料理アシスタントを開発しました！ 入賞は叶いませんでしたが「ハッカソンはプロダクト開発とは違う」「可能性の余地を残しておく」「あらかじめ時間の割り振りを決めておく」「評価基準を鵜呑みにしない」など学べることがたくさんありました また「ぬこぬこさんのポストを見て申し込みました！」と伝えてくださる方がなぜか会場にたくさんいらして、微力ながら良い影響力の使い方というのを学びました笑 ハッカソン外で作ったものですが、私は私で Voxtral を用いた自作の音声入力アプリを日常的に使っているので、こちらも近いうちに公開できればと整備を進めています！Voxtral はいいぞ！ Mistral AI のウェブチャットの Le Chat はフランス語で猫の意味、会場に最後まで残っていたらレゴをもらってしまったので早速、夜な夜な作った↓

Mistral Worldwide Hackathon で私を倒しませんか？笑 世界 7 都市同時開催なので、手を動かしながら世界中でどんなものが作られ、どう評価されるかの知見が一気に得られるハッカソンです！一緒に参加しましょ！Mistral AI の人曰く、日本人が多くないので、通りやすいはず！ luma.com/mistralhack-to…

Introducing Mistral AI's biggest hackathon ever! 📅 Feb 28 - Mar 1 🌍 Paris | London | NY | SF | Tokyo | Singapore | Sydney & online 48 hours. The best hackers. 🤝 Partners: @wandb @nvidia @awscloud @HackIterate 🏆 $200K in prizes. Special awards from @elevenlabs @huggingface @JUmp @whitecircle @supercell Link in 🧵

Great app: Voxtral-Subtitles to transcribe any video with word-level subtitles, speaker diarization & multilingual translation. ⬇️ Try in now on Hugging Face Spaces

@HowDevelop @BleedingDev @MistralAIDevs @xenovacom Already working together on this 🤗

@BleedingDev @MistralAIDevs @xenovacom would love to see the port for this via Tranformers.js!

Voxtral can now directly stream audio input into text output. Perfect for: - Live subtitles - Language learning apps - Note-taking tools - And more! Made a demo for you to try directly on hugging face !

@BleedingDev @MistralAIDevs @HowDevelop Very soon 👀The model is small, oss and we'll make a WebGPU version

@MistralAIDevs @HowDevelop wonder how soon till we can run this in browser!

Automate or simplify your workflow for subtitle creation with Voxtral Transcribe 2.0! From word-level timestamps to diarization and context biasing, take it further by leveraging our other general-purpose models to translate subtitles.

@drivelinekyle @simonw Sorry it didn’t work! The demo was set up for ~50 concurrent users (backed by a VLLM endpoint), but we got thousands. Switched to our API for now and will add a local option soon when the model is supported in transformers.js! Give it another try!

@simonw press X to doubt

The demo on huggingface.co/spaces/mistral… is worth a try - ignore the "No microphone found" message, clicking "Record" and allowing your browser to use a microphone fixes that. It transcribes very accurately in almost real-time. It's really impressive.

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! 🤯

ヨーロッパ発LLM ProviderであるMistral AIから最新リリース×日本企業での活用法をライブデモで紹介してくれることになりました💻 LLM Providerが続々と日本のAI市場を意識されている流れは興味深いです @Jthmas404 @MistralAI fullyconnected.jp #mistral

Magistral from @MistralAI is a model trained to do reasoning on images. Here is a fun challenge : Can you bet Magistral 1.2 at Geoguessr ? 🌎❓ To test that, I recreated an MVP on @huggingface 🔗 below

𝗥𝗲𝗴𝗶𝘀𝘁𝗲𝗿 𝗵𝗲𝗿𝗲 : huggingface.co/Agents-MCP-Hac… Live : youtube.com/watch?v=MU7Fyx…

Mistral AI is sponsoring the @huggingface 𝗚𝗿𝗮𝗱𝗶𝗼 𝗔𝗴𝗲𝗻𝘁𝘀 & 𝗠𝗖𝗣 𝗛𝗮𝗰𝗸𝗮𝘁𝗵𝗼𝗻 𝟮𝟬𝟮𝟱 Take this opportunity to explore our new 𝗠𝗖𝗣 𝗰𝗼𝗺𝗽𝗮𝘁𝗶𝗯𝗹𝗲 𝗔𝗴𝗲𝗻𝘁𝘀 𝗔𝗣𝗜. Join me for the 𝗸𝗶𝗰𝗸𝗼𝗳𝗳 𝗹𝗶𝘃𝗲 tomorrow (June 3rd)! Details in thread

@tomaarsen @MistralAI @OpenAI @GoogleDeepMind here 👉 huggingface.co/spaces/Jofthom… 1) Click on choose name 2) Click Find a user and search for the currently active agent displayed on screen Agent names are : OpenAIAgent, MistralAgent and GeminiAgent 3) Send the invitation ! 4) Fight! 💥🕹️

@Jthmas404 @MistralAI @OpenAI @GoogleDeepMind Looks awesome! How can the livestream be used? I see that the Agent is Ready right now

🎶 Wanna be the very best... Pokémon agent coder? 🎮 New Agents Course bonus unit! Learn function calling to build agents & challenge @MistralAI / @OpenAI / @GoogleDeepMind strategies! LIVE happening now + leaderboard soon! 🔗 below