Google launches Gemma 4 12B, a unified encoder-free multimodal model

LaunchAI Models

8 days ago

Google launches Gemma 4 12B, a unified encoder-free multimodal model

The 12B parameter model runs on laptops with 16GB VRAM, supports native audio inputs, and approaches the performance of the larger 26B MoE version. It is released under Apache 2.0 and includes multi-token prediction for reduced latency.

Gemma 4 QAT models: Optimizing model compression for mobile and laptop efficiency12 days agoOmar Sanseviero

Bringing Gemma 4 12B to your Laptop: Unlocking Local, Agentic Workflows with Google AI Edge14 days ago

Gemma 4 12B: The Developer Guide14 days ago

OBLITERATUS/Gemma-4-12B-OBLITERATED9 days agoOBLITERATUS

yuxinlu1/gemma-4-12B-coder-fable5-composer2.5-v1-GGUF4 days agoyuxinlu1

google/gemma-4-12B-it-qat-q4_0-gguf11 days agogoogle

unsloth/gemma-4-12B-it-qat-GGUF12 days agounsloth

google/gemma-4-12B14 days agogoogle

google/gemma-4-12B-it-assistant13 days agogoogle

google/gemma-4-12B-it14 days agogoogle

RT @osanseviero: Gemma 4 with 255 tokens per second running directly in your browser...12 hours agoHugging Face

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

RT @UnslothAI: Google releases Gemma 4 QAT. ✨ You can now run Gemma 4 at 3x less memory with near...12 days agoHugging Face

RT @googlegemma: We just dropped Gemma 4 Quantization-Aware Training (QAT) checkpoints on Hugging...12 days agoHugging Face

8 days ago