Google DeepMind Introduces Gemma 4 12B Multimodal Model

Google DeepMind has officially introduced its latest model, designated Gemma 4 12B. The company describes this new system as a unified, encoder-free multimodal model.

This announcement highlights a new development from Google DeepMind, an artificial intelligence research laboratory.

According to the Google DeepMind Blog, the introduction of Gemma 4 12B marks a step forward in their ongoing work with AI models.