Google unveils Gemma 4 12B: encoder-free multimodal model for laptops

Google Blog (General)

Gemma 4 12B natively processes audio and images without separate encoders, runs on 16GB RAM. It is Google's first mid-sized model with native audio inputs, released under Apache 2.0 license.

GemmaMultimodal

See 14 more sources

Bringing Gemma 4 12B to your Laptop: Unlocking Local, Agentic Workflows with Google AI Edge16 days ago

Introducing Gemma 4 12B: a unified, encoder-free multimodal model16 days agoGus Martins

Gemma 4 12B: The Developer Guide16 days ago

OBLITERATUS/Gemma-4-12B-OBLITERATED11 days agoOBLITERATUS

yuxinlu1/gemma-4-12B-coder-fable5-composer2.5-v1-GGUF6 days agoyuxinlu1

google/gemma-4-12B-it-qat-q4_0-gguf13 days agogoogle

unsloth/gemma-4-12B-it-qat-GGUF14 days agounsloth

google/gemma-4-12B16 days agogoogle

google/gemma-4-12B-it-assistant15 days agogoogle

google/gemma-4-12B-it16 days agogoogle

RT @AndreasPSteiner: Released last week, and already more than 4M downloads on HuggingFace alone 😊...6 days agoHugging Face

RT @googlegemma: Meet Gemma 4 12B! A unified, encoder-free multimodal model designed to bring...16 days agoHugging Face

RT @UnslothAI: Google releases Gemma 4 QAT. ✨ You can now run Gemma 4 at 3x less memory with near...14 days agoHugging Face

RT @osanseviero: Gemma 4 with 255 tokens per second running directly in your browser...2 days agoHugging Face

Google Blog (General)·Omar Sanseviero

Jun 3, 4:00 PM