Skip to main content
Measured savings across 11 LLMs, from Claude Opus 4.7 to Gemini Flash.→ See per-model data
Connect your client
Models

Google releases DiffusionGemma as open-weight model with 500+ tokens per second

Google open-sourced DiffusionGemma, a 26B parameter Gemma model under Apache 2 license, delivering 500+ tokens/second throughput on NVIDIA's NIM cloud API.

1 min read

Google has released DiffusionGemma as an open-weight model, making available the research behind its experimental Gemini Diffusion model from May 2025. The new model, called google/diffusiongemma-26B-A4B-it, is licensed under Apache 2 and available on Hugging Face. This marks a shift from the compan...

Sign in to read the full analysis

Free account. Full analysis on LLM unit economics, plus the weekly Cost-of-Inference column.

Method & sources
Source type
Primary publication (lab/vendor blog) — our analysis + implication
Source link
Simon Willison
Published
UTC
Byline
By the gotcontext.ai team (editorial standards)
Correction?
corrections@gotcontext.ai