Sabitlenmiş Tweet
Clément Farabet
2.6K posts
Clément Farabet
@clmt
AI @ Google DeepMind (Gemini, Gemma & Beyond). Ex NVIDIA (self-driving cars, https://t.co/QtrCBg3wx0), Twitter (founded Cortex), MadBits (founded+sold) 🇺🇸🇫🇷
London + Palo Alto Katılım Aralık 2009
2.4K Takip Edilen19.4K Takipçiler
Clément Farabet retweetledi
Really cool work from the team reimagining the mouse pointer to be intelligent! Try the prototype in @GoogleAIStudio it's pretty magical.
Google DeepMind@GoogleDeepMind
We’re reimagining a 50-year-old interface - the mouse pointer - with AI. 🖱️ These experimental demos show how people can intuitively direct Gemini on their screens using motion, speech, and natural shorthand to get things done 🧵
English
Clément Farabet retweetledi
Gemma 4 just got a massive speed upgrade! ⚡️🏎️💥
Google just released Multi-Token Prediction (MTP) drafters that deliver up to a 3x faster inference boost!
💬 Super fast chat & low latency voice on small models 🎙️
📱 Faster on-device edge hardware performance 💻
🧠 Same frontier-class reasoning, a fraction of the wait ⏳
English
Clément Farabet retweetledi
Gemma 4 just got even faster!
We're releasing Multi-Token Prediction (MTP) drafters that deliver up to a 3x speedup, without any degradation in output quality or reasoning logic.
GIF
English
Google AI Developers@googleaidevs
Speed up your Gemma 4 workflows by up to 3x with Multi-Token Prediction (MTP) drafters. Standard LLM inference is fundamentally memory-bandwidth bound, creating a latency bottleneck as billions of parameters travel from VRAM just to generate a single token. We're working to ease this bottleneck with MTP drafters for @googlegemma 4.
QHT
Clément Farabet retweetledi
Incredibly proud to see Alphabet recognized in the 2026 @TIME 100 Most Influential Companies!
Huge congratulations to all the teams across @Google working hard to turn AI breakthroughs into everyday reality for our users! ✨ #TIME100Companies
News from Google@NewsFromGoogle
“In 2016, [@SundarPichai] had declared Google would be an “AI-first company,” and began cultivating a series of projects—custom chips, Cloud, YouTube, and deep AI research—that seemed to have nothing to do with Google’s core search product. All of these bets have paid off, and then some.” — @TIME in their new #TIME100Companies cover story on Alphabet and AI → time.com/collection/tim… 📷: Daniel Dorsa
English
@clmt thank you! Im spending every evening working on it. Having a lot of fun, hope folks will be inspired to bring local models into their lives.
English
Amazing project
Dmitry Lyalin@LyalinDotCom
I've been working on a side project for the last few weeks... what if you could have a Gemma powered app that would let you have a personal assistant that could browse the internet with you, do research, code, etc Here's my MacOS experiment github.com/LyalinDotCom/G… More info 👇
English
Clément Farabet retweetledi
What does it take to run 3, 5, or even 10 concurrent instances of Gemma 4 locally?
We've open-sourced a demo letting you run multiple models side-by-side on your hardware.
Gemma 4 26B A4B easily runs 10+ concurrent requests on a MacBook Pro M4 Max at 18 tokens/sec per request.
English
Clément Farabet retweetledi
We are launching two powerful updates to Deep Research in the Gemini API, now with better quality, MCP support, and native chart/infographics generation.
Use Deep Research when you want speed and efficiency, and use Max when you want the highest quality context gathering & synthesis using extended test-time compute — achieving 93.3% on DeepSearchQA and 54.6% on HLE.
English
Clément Farabet retweetledi
Our most expressive and steerable TTS model yet! Designed to give builders granular control over AI-generated speech, Gemini 3.1 Flash TTS is really fun to play with! Available in preview today - for devs via the Gemini API & @GoogleAIStudio + for enterprises on Vertex AI
Logan Kilpatrick@OfficialLoganK
Introducing Gemini 3.1 Flash TTS 🗣️, our latest text to speech model with scene direction, speaker level specificity, audio tags, more natural + expressive voices, and support for 70 different languages. Available via our new audio playground in AI Studio and in the Gemini API!
English
Clément Farabet retweetledi
Gemma 4 looks at a parking lot. Decides what to ask. Calls SAM 3.1.
"Segment all vehicles." 64 found.
"Now just the white ones." 23 found.
One model reasoning and orchestrating. One model executing.
Both running locally on a MacBook. MLX. No cloud. No API.
English
Clément Farabet retweetledi
We’re taking Gemma 4 on a world tour with 50+ events across the globe! From hackathons to deep dives, come build with the latest open weights in a city near you 🌎
Where we’ll be next: San Francisco, São Paulo, Kolkata, Las Vegas, Seoul, online, and more!
English
Clément Farabet retweetledi
Lots of love for Gemma 4! Team just told me it’s already had 10M+ downloads since last week’s launch. Gemma models have now been downloaded 500M+ times! Excited to see what you all are creating 👀
English
Clément Farabet retweetledi
Gemma 4 models are now on Mac!
Try the new Gemma 4 E2B and E4B — Google’s most intelligent open models for the edge, powered by MLX for best-in-class performance on M-series chips.
Update your Mac app now.
English
Clément Farabet retweetledi
Clément Farabet retweetledi
Gemma 4 can run on phones without an internet connection! 🤯
It can perform local agentic tasks, such as logging and analyzing trends. When connected, it can also make API calls.
Want to try it yourself? Get the Google AI Edge App on iOS or Android. (🔊 Sound on for the demo!)
English
Clément Farabet retweetledi
Gemma 4 26B-A4B is now ~2x faster at 375K context with TurboQuant on MLX-VLM v0.4.4 🚀
The model's official max context is 262K but I pushed it to 375K anyway. That's roughly 5–6 full novels (the entire LOTR trilogy + The Hobbit).
Up to ~20K tokens they're neck and neck, but after that TBQ dominates with ~1GB memory savings.
KV savings are modest (4–17%) because only 5/30 layers get compressed. But those 5 layers dominate decode time at long contexts, so the speed gains are massive.
Device: M3 Max 96GB
English
Clément Farabet retweetledi
Gemma 4 from @GoogleDeepMind has hit 2.5B tokens so far:
Two flavors: 31B dense, & 26B MoE (running 5x faster than the dense version now!)
256K context • native function calling • multimodal • configurable thinking • 140+ languages
Try the 26B MoE: openrouter.ai/google/gemma-4…
English