Google DeepMind releases DiffusionGemma for 4x faster parallel text generation.

Google DeepMind released DiffusionGemma, an experimental open model that generates multiple words in parallel rather than sequentially, enabling low-latency text generation for single-user workloads. NVIDIA optimized the model to run faster across GeForce RTX GPUs, RTX PRO platforms, and DGX Spark systems from local PCs to the cloud.

Sources

X mentions

13k ▲

First seen

22Dago

Velocity

+11%/6h

CONTRIBUTING SOURCES

6 ARTICLES

X DISCOURSE

13k TOTAL · TOP 3

@demishassabis3H · 300.7K

I’ve always believed the No.1 application of AI should be to improve human health. That work started with AlphaFold, and now at @IsomorphicLabs with the mission to reimagine drug discovery and one day solve all disease! We are turbocharging that goal with $2.1B in new funding. https://t.co/Hvk20dHgjl

@demishassabis23H · 299.4K

@demishassabis2D · 295.9K