Google's open-source diffusion language model generates 256 tokens in parallel and self-corrects, hitting 4x speed on one GPU at a cost to quality.

Deploying DFlash block diffusion on NVIDIA hardware accelerates autoregressive LLMs during latency-sensitive inference.

Google LLC today released DiffusionGemma, a large language model based on an emerging machine learning approach known as text diffusion. The company says the algorithm can generate text four times ...

The boffins on Google’s DeepMind team unveiled an experimental new language model this week that uses techniques originally developed for AI image generators to boost text output performance by as ...

Customer stories Events & webinars Ebooks & reports Business insights GitHub Skills ...

[CVPR 2026] Language-Guided One-Step Diffusion Model for Nighttime Flare Removal [ICML 2026] PODiff: Latent Diffusion in Proper Orthogonal Decomposition Space for Scientific Super-Resolution ...

Two systems with identical parameter counts can behave dramatically differently depending on how they are built.

A drop of dye added to a glass of water undergoes ordinary diffusion. However, when placed on the surface of a foam, the dye spreads differently—diffusion becomes anomalous. An example of this is the ...

Another day, another AI model from Google. This time, Google DeepMind has released a new member of the Gemma 4 open model family, but it’s fundamentally different from the rest of the lineup.

Rather than generating text word by word, Google's experimental open-source model drafts entire passages simultaneously using diffusion, resulting in up to 4x faster inference. Extremely powerful ...