AI updates
Gemma 4 12B: an encoder-free multimodal model for laptops
Google introduced Gemma 4 12B, a mid-sized open model where vision and audio inputs flow straight into the LLM backbone with no separate encoders; it runs in 16GB and ships with multi-token-prediction drafters.
- No separate vision or audio encoder: image and audio inputs go directly into the language model.
- It runs locally with 16GB of VRAM or unified memory, under Apache 2.0.