Introducing Gemma 4 12B: a unified, encoder-free multimodal model

# What Google Just Released and Why It Matters Google DeepMind built a new AI model called Gemma 4 12B that can understand both text and images in a single, streamlined system—kind of like giving an AI assistant both eyes and ears at the same time. The "12B" means it's relatively compact and efficient, so companies can run it on their own computers without needing massive data centers, making powerful AI more accessible and affordable. This matters because it lowers the barrier for businesses to add smart image-and-text understanding to their apps without paying hefty fees to use someone else's service.