Google 把去年惊艳的扩散模型开源了，做图像/文本生成的开发者可以直接用 NVIDIA 免费 API 体验，速度超 500 tokens/秒，值得立刻上手试。
Google 去年曾短暂发布实验性的 Gemini Diffusion 模型，如今以开源形式回归，推出 DiffusionGemma-26B-A4B-it 模型，采用 Apache 2 许可证。该模型在 NVIDIA NIM 云 API 上免费托管，生成速度可达 500+ tokens/秒。作者实测生成 2409 tokens 仅需 4.4 秒，性能出色。这是 Google 在扩散模型领域的重要开源动作，为开发者提供了高性能的生成式 AI 选项。
DiffusionGemma
DiffusionGemma Last May Google briefly released an experimental Gemini Diffusion model. I tried the preview at the time and recorded it running at 857 tokens/second. It was an exciting model, but Google made no further announcements about it. That research has returned in the best possible way: as a new open weight (Apache 2 licensed) Gemma model, google/diffusiongemma-26B-A4B-it . NVIDIA are currently hosting the model for free on their NIM cloud API. I used that API to generate this pelican , which took 4.4s (according to time uv run generate.py ) to return 2,409 tokens - so at least 500 tokens/second. Via Hacker News Tags: google , ai , generative-ai , llms , nvidia , pelican-riding-a-bicycle , gemma , llm-release , llm-performance