SenseNova-U1 open-sources 8-step distilled LoRA, greatly improving inference speed
Original title:🚀 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮-𝗨𝟭 𝘂𝗽𝗱𝗮𝘁𝗲: ⚡ 𝗢𝗽𝗲𝗻-𝘀𝗼𝘂𝗿𝗰𝗲𝗱 𝗮𝗻 𝟴-𝘀𝘁𝗲𝗽 𝗱𝗶𝘀𝘁𝗶𝗹𝗹�...
AI summary
🚀 SenseNova-U1 update: ⚡ Open-sourced 8-step distilled LoRA: reduced from 100 NFE to 8 NFE, with H100 inference time shortened from 23 seconds to 2 seconds 🧩 Now supports ComfyUI, providing ready-to-use workflows for text-to-image, image editing, and interleaved generation Try it here 👇 https://github.com/OpenSenseNova/SenseNova-U1/
Key Takeaways
- 🚀 SenseNova-U1 update: ⚡ Open-sourced 8-step distilled LoRA: reduced from 100 NFE to 8 NFE
- with H100 inference time shortened from 23 seconds to 2 seconds 🧩 Now supports ComfyUI
- providing ready-to-use workflows for text-to-image,image editing,and interleaved generation Try it…
Quick Facts
- Published
- Updated
- Category
- Model Releases
- AI score
- 71
从100步到8步,23秒压到2秒,商汤这个蒸馏LoRA把U1的推理成本打下来了,做实时图像应用的可以认真看看,ComfyUI一接就能跑。
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: SenseNova-U1 open-sources 8-step distilled LoRA, greatly improving inference speed. May 6, 2026. https://new.nat6.net/en/news/cmou5uboa00l4slndg69l7v7x/