bytedance/USO

View on GitHub

[CVPR 2026] 🔥🔥 Official Repo of USO: Unified Style and Subject-Driven Generation via Disentangled and Reward Learning

Stars: 1.2k
Forks: 77
Open beginner issues: 0
Indexed issues: 26

Dominant language: Python
License: Apache-2.0
Last GitHub push: Sep 12, 2025
Latest indexed: Sep 14, 2026
Contributing guide: No contributing guide
Code of conduct: No code of conduct
Beginner labels: No beginner labels indexed
PR merge metrics: No merged PRs in 30d

Open All

26 open issues indexed

Questions about the training process Open

bytedance/USO#2 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
quantized code/checkpoint Open

bytedance/USO#5 · 7 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
License Open

bytedance/USO#8 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Blend character with scene Open

bytedance/USO#9 · 5 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Preserving Identify? Open

bytedance/USO#10 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Release USO-Bench dataset on Hugging Face Open

bytedance/USO#12 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
关于人物一致性的问题 Open

bytedance/USO#13 · 4 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
【ComfyUI已支持】lora训练+comfyui支持 Open

bytedance/USO#14 · 18 comments · 2 reactions · 0 assignees · Indexed Sep 14, 2026
Style-subject driven generation for already stylized input images Open

bytedance/USO#17 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Is flux-dev the only supported base model? Open

bytedance/USO#19 · 3 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
macOS MPS: inference produces noisy output and setup issues — investigation and suggested fixes Open

bytedance/USO#20 · 3 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
面部崩坏 Open

bytedance/USO#23 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Controlnet support Open

bytedance/USO#24 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Error when starting app.py Open

bytedance/USO#26 · 11 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
什么时候release训练代码呢 Open

bytedance/USO#32 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Plans to integrate this for Qwen-Image too ? Open

bytedance/USO#33 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
one-click setup一键本地部署 Open

bytedance/USO#35 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Feature Request: Batch Inference Support in Gradio App Open

bytedance/USO#36 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
云端免费体验bug欢迎报给我谢谢需要48gb显存运行 Open

bytedance/USO#37 · 0 comments · 1 reaction · 0 assignees · Indexed Sep 14, 2026
Can I build a style image library with pooled SigCLIP features for more robust and generalizable style transfer? Open

bytedance/USO#38 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
After some testing, I found that the style of the original picture can be changed by referring to the reference style picture, but it is difficult to change the original character's posture and movement in combination with the prompt word. Open

bytedance/USO#39 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
USO + ControlNet (Depth/Canny/union) with Flux-1-Dev to place a subject into a fixed background with minimal changes Open

bytedance/USO#40 · 2 comments · 5 reactions · 0 assignees · Indexed Sep 14, 2026
when release datasets and training code？look forward Open

bytedance/USO#41 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
VLMs for data generation Open

bytedance/USO#42 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Release the code of CSD score? Open

bytedance/USO#43 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
About the training of the first stage Open

bytedance/USO#45 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026

bytedance/USO

Get new issues in your inbox