🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Really excited that we can finally share Gemma 3 with the world. The whole team spent a lot of hard work on this and the results speak for themselves: Being able to fit a top 10 LMSys model on a single accelerator will enable so many people to benefit from strong models.

Checkout our latest Gemma models, you'll love them :) Also checkout the results on our BIG-Bench Extra Hard benchmark:

Congratulations to the whole Gemma team for the launch and especially Aishwarya Kamath who did an amazing job pushing the MM capability of the model 🚀. Give a try to the model 🔥

Omar Sanseviero okay this is dope af

We want to make it easier to sample/finetune Gemma models. I have watched the insanely talented github.com/Conchylicultor build this in the last few months. Feedback appreciated as we are looking to improve the library!

Hiring two student researchers for Gemma post-training team at Google DeepMind Paris! First topic is about diversity in RL for LLMs (merging, generalization, exploration & creativity), second is about distillation (with Nino Vieillard). Ideal if you're finishing PhD. DMs open!

It is very hard to find the right balance between usability and performance to help the developer community. For Gemma, we spent time asking for feedback from the community and built based on this feedback. We ended up entirely focused on models at useful sizes, i.e. download any

Gemma keeps delivering! I'm very excited to share with you the most recent LMArena results for the Gemma 3 family💥 Gemma 3 stays as the best open model that can run on a single GPU. And stay tuned, more to come!

I've been reading about Gemma 3n for months. It sounded great, but my mind was blown when I started seeing the demos. 🤯 Check out this quick demo:

Releasing Gemma 3n, our new open-weight model processing audio, images and text (with improved multilingual capabilities), optimized for on-device usage with MatFormer architecture (enabling adaptive compute) and reaching 1283 on Chatbot Arena. Read more: developers.googleblog.com/en/introducing….

Gemini Nano meets Gemma! Gemma 3n the next generation of Gemini Nano is expanding to multimodality for edge devices! ✨ Gemma 3n will be an open, offline first model to run and build agents from browsers to on-device! 🚀 Gemma 3n will: 🔤 👀 🖼️ Understand text, images and audio

Gemini Diffusion is out! Very excited to have worked on the post-training of such a state-of-the-art text diffusion model. Incredible performance at lightspeed⚡️ Congrats to everyone involved!!

Pocket powerhouse admist I/O awesomeness! Gemma 3n E4B & E2B are insane models, optimized for on-device while rivaling frontier models. It's a 🪆Matryoshka Transformer (MatFormer)🪆: Natively elastic b/w 4B & 2B pareto-optimally! ⭐️: free models with ZERO training cost! 🧵👇

Really proud that two new models have been presented at I/O which we have post-trained: - Gemini Diffusion: with >1k tokens per second a completely new LLM experience deepmind.google/models/gemini-… - Gemma 3n: pushing the boundary of what is possible on mobile developers.googleblog.com/en/introducing…

We just released Gemma 3n, a mobile-first & multimodal LLM that works with as little as 2Gb RAM. Feels crazy to interact with a model whose training I contributed to, hosted on my *own* phone (see screenshot!) 🤯 It packs so much for its size, give it a try (how to in thread)!

This is my favorite demo of Gemma 3n -- multimodal live video understanding and intelligence, locally on your phone 🤯! This was only possible with the peak of foundation models at I/O last year -- the Astra demo -- the progress of small models is incredible

Unmute is our new cascaded voice assistant: fast, accurate, and flexible. It doesn't have the full-duplex and zero latency of Moshi, but you can change the voice with a 10s sample and plug any LLM. A good playground for testing custom voice AIs.

Everyone is talking about how we need more AI data centers (especially the ones who would mostly benefit from them) but why is no one talking about on-device AI? Running AI on your device: - Free - Faster & takes advantage of existing hardware - 100% privacy and control (you