Google Open-Sources Gemma 4 12B: Encoder-Free Multimodal for 16GB VRAM

📡 AI News 💬
Google Open-Sources Gemma 4 12B: Encoder-Free Multimodal for 16GB VRAM

🩺 Summary

Google released Gemma 4 12B, a dense multimodal model running on 16GB VRAM without encoders.

📝 Details

Gemma 4 12B skips vision/audio encoders, feeding data directly into the LLM. 256K context, Apache 2.0. Designed for laptops with 16GB VRAM.
Gemma 4 12B skips vision/audio encoders, feeding data directly into the LLM. 256K context, Apache 2.0. Designed for laptops with 16GB VRAM.