Introducing Gemma 4 12B: a unified, encoder-free multimodal model
Get smart on it
Gemma 4 12B is a new multimodal AI model designed to run directly on laptops, handling text, images, and audio inputs without requiring separate encoding systems. It differs from traditional multimodal models by processing visual and audio data through a unified architecture that feeds directly into the language model, reducing memory requirements and processing latency. The model performs comparably to a larger 26B model while requiring less than half the memory footprint, making it capable of running on consumer laptops with 16GB of RAM. It is being released under an open license with support across multiple developer tools and frameworks.
Related news
Vibe coding platform Base44 launches own model as AI startups seek defensibility
Wix-owned vibe coding platform Base44 has started rolling out its own AI model — with hopes that it will eventually outperform frontier models.
DiScoFormer: One transformer for density and score, across distributions
DiScoFormer is a transformer model that estimates both the density and score of a data distribution in a single forward pass without requiring retraining for new distributions. The density describes where data points cluster, while the score, the gradient of log-density, points toward more probable regions and is used in diffusion-based generative models, Bayesian sampling, and scientific simulations. Existing methods force a trade-off: classical kernel density estimation works on any distribution but loses accuracy in high dimensions, while neural score-matching models stay accurate in high dimensions but must be retrained for each new distribution. DiScoFormer significantly outperforms kernel density estimation, cutting score error by 6.5 times and density error by more than 37 times in 100 dimensions, while generalizing to distributions with shapes and complexities not seen during training.
