Thomas Mesnard

🚀 After Gemma 3 1B, we’re going tiny: Gemma 270M — fast, on-device, low-power & privacy-first. A new vision: smaller models with strong instruction-following & finetuning, ideal for low-latency edge apps & automation. Congrats to everyone involved! huggingface.co/google/gemma-3…

An exciting milestone for AI in science: Our C2S-Scale 27B foundation model, built with @Yale and based on Gemma, generated a novel hypothesis about cancer cellular behavior, which scientists experimentally validated in living cells.  With more preclinical and clinical tests, this discovery may reveal a promising new pathway for developing therapies to fight cancer.

@giffmana 🔥

@Mesnard_Thomas right with gemma3 it's all but two of this screen lol

Nice! Also Gemma and in particular Gemma 3 1B 👨‍🍳🤌😍

Thrilled to announce the world’s most capable differentially private LLM! Huge congratulations to the entire team — and special kudos to Amer Sinha for his outstanding contributions. Take a look! services.google.com/fh/files/blogs…

Introducing EmbeddingGemma, our newest open model that can run completely on-device. It's the top model under 500M parameters on the MTEB benchmark and comparable to models nearly 2x its size – enabling state-of-the-art embeddings for search, retrieval + more.

Introducing EmbeddingGemma🎉 🔥With only 308M params, this is the top open model under 500M 🌏Trained on 100+ languages 🪆Flexible embeddings (768 to 128 dims) with Matryoshka 🤗Works with your favorite open tools 🤏Runs with as little as 200MB developers.googleblog.com/en/introducing…

We've had a busy August!

Looking for a small or medium sized VLM? PaliGemma 2 spans more than 150x of compute! Not sure yet if you want to invest the time 🪄finetuning🪄 on your data? Give it a try with our ready-to-use "mix" checkpoints: 🤗 huggingface.co/blog/paligemma… 🎤 developers.googleblog.com/en/introducing…

Thrilled to launch the PaliGemma 2 Mix models today! Try out class-leading vision-language models from 3B to 28B =). #gemmaverse

🥳🥳🥳

Leaving X for Bluesky ! bsky.app/profile/tmesna…

🚀🚀PaliGemma 2 is our updated and improved PaliGemma release using the Gemma 2 models and providing new pre-trained checkpoints for the full cross product of {224px,448px,896px} resolutions and {3B,10B,28B} model sizes. 1/7

Thrilled to announce the launch of PaliGemma 2, a state-of-the-art update to our PaliGemma vision-language models! These models are based on Gemma 2, and are available today in 2B, 9B, and 27B sizes. developers.googleblog.com/en/introducing… huggingface.co/blog/paligemma2

📢 Exciting opportunity alert! We (valeo.ai) just posted our annual research internship openings in computer vision & ML. Check out the openings and the great achievements by our past interns here: valeoai.github.io/interns/

Gemma 2 just got even better! 🚀 New Japanese-tuned 2B model AND a $150K Kaggle competition to build Gemma models for every language. Great to have @sundarpichai here to share the excitement! Read more: goo.gle/Gemma4Japan #GemmaDeveloperDay

❤️

Bravo!

Princeton's Gemma 2 9B SimPO fine-tuned model is doing amazing on @lmsysorg 🤯📈 You can find it on @huggingface : huggingface.co/princeton-nlp/… Nice work @yumeng0818 @xiamengzhou & @danqi_chen

Imagine a 27B LLM can beat a 405B model in writing quality code by investing a few milliseconds in static code repair. Now stop imagining and take a look at this chart 🌈 Just for Go, we have the following stats: - Increases score +22.9% across 45 applicable models - +26.2% response files compiled (avg. 17 files, 150 tasks total) - mistral-tiny has +71% in score: beats mistral-small and mistral-medium - Gemma 2 27B has +16% in score: beats GPT4o and Llama 3.1 405B Proof that the approach of doing code fixes over a static analysis should be the default for every code response and coding assistant. Makes code instantly more useful.