Google's DiffusionGemma AI Hits 1,000 Tokens Per Second—And It's Free

Crypto Insight Summary: Image: Shutterstock/Decrypt]( Google's DiffusionGemma AI Hits 1,000 Tokens Per Second—And It's Free * * * By Edited by Jun 10, 2026 Jun 10, 2026 4 min read Google. Image: Shutterstock/Decrypt Create an account to save your articles. * Google released DiffusionGemma, a free open-weight model that generates entire 256-token blocks simultaneously via text diffusion—hitting over 1,000 tokens per second on an NVIDIA H100, four times faster than standard autoregressive models. * On NVIDIA NIM, the model arrived preconfigured at 8,192 tokens of context—below the 64,000-token floor that agentic frameworks like Hermes Agent require—meaning autonomous workflows won't run without manual reconfiguration. Google dropped DiffusionGemma , an open model AI that generates text the way image generators create pictures: start with noise, refine until it makes sense. It hits 1,000 tokens per second on an NVIDIA H100. It’s also free, Apache 2.0, with weights on Hugging Face. Per , the model hits "700+ tokens per second on NVIDIA GeForce RTX 5090." It also trails standard Gemma 4 on output quality. Crypto Insight Analysis: Direction: Neutral Impact: Minor Impact Horizon: Short-term (1-3 days) Reasoning: AI advancements typically have limited direct impact on crypto markets unless tied to specific blockchain integration.