Upload 89 files

.gitattributes

@@ -33,3 +33,32 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+assets/comfyui_template.png filter=lfs diff=lfs merge=lfs -text
+assets/gradio_examples/4subject/ref.jpg filter=lfs diff=lfs merge=lfs -text
+assets/gradio_examples/identity1_result.png filter=lfs diff=lfs merge=lfs -text
+assets/gradio_examples/style3_style4_result.webp filter=lfs diff=lfs merge=lfs -text
+assets/gradio_examples/style3.webp filter=lfs diff=lfs merge=lfs -text
+assets/gradio_examples/style4.webp filter=lfs diff=lfs merge=lfs -text
+assets/gradio_examples/z1_mix_style/ref1.webp filter=lfs diff=lfs merge=lfs -text
+assets/gradio_examples/z1_mix_style/ref2.webp filter=lfs diff=lfs merge=lfs -text
+assets/gradio_examples/z3_mix_style/ref1.jpg filter=lfs diff=lfs merge=lfs -text
+assets/show_case1.webp filter=lfs diff=lfs merge=lfs -text
+assets/show_case2.webp filter=lfs diff=lfs merge=lfs -text
+assets/show_case3.webp filter=lfs diff=lfs merge=lfs -text
+assets/show_case4.webp filter=lfs diff=lfs merge=lfs -text
+assets/show_case5.webp filter=lfs diff=lfs merge=lfs -text
+assets/show_case6.webp filter=lfs diff=lfs merge=lfs -text
+assets/show_case7.webp filter=lfs diff=lfs merge=lfs -text
+assets/show_case8.webp filter=lfs diff=lfs merge=lfs -text
+assets/teaser.webp filter=lfs diff=lfs merge=lfs -text
+assets/usoxcomfyui_official.jpeg filter=lfs diff=lfs merge=lfs -text
+assets/usoxcomfyui.webp filter=lfs diff=lfs merge=lfs -text
+workflow/17-17-29.webp.webp filter=lfs diff=lfs merge=lfs -text
+workflow/example1.png filter=lfs diff=lfs merge=lfs -text
+workflow/example2.png filter=lfs diff=lfs merge=lfs -text
+workflow/example3.png filter=lfs diff=lfs merge=lfs -text
+workflow/example4.png filter=lfs diff=lfs merge=lfs -text
+workflow/example5.png filter=lfs diff=lfs merge=lfs -text
+workflow/example6.png filter=lfs diff=lfs merge=lfs -text
+workflow/input.png filter=lfs diff=lfs merge=lfs -text
+workflow/style5_0.webp.webp filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,195 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# UV
+#   Similar to Pipfile.lock, it is generally recommended to include uv.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#uv.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/latest/usage/project/#working-with-version-control
+.pdm.toml
+.pdm-python
+.pdm-build/
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/
+
+# Ruff stuff:
+.ruff_cache/
+
+# PyPI configuration file
+.pypirc
+
+# User config files
+.vscode/
+output/
+
+# ckpt
+*.bin
+*.pt
+*.pth
+
+logs/
+*.safetensors
+
+
+
+# FOR EXCLUSION OF DOWNLOADED WEIGHTS
+# Ignore everything in weights/
+weights/*
+
+# But don't ignore this file
+!weights/downloader.py

LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

README.md

@@ -1,13 +1,243 @@
----
-title: Video Generator
-emoji: 🐠
-colorFrom: indigo
-colorTo: purple
-sdk: gradio
-sdk_version: 5.44.1
-app_file: app.py
-pinned: false
-license: mit
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+<h3 align="center">
+    <img src="assets/uso.webp" alt="Logo" style="vertical-align: middle; width: 95px; height: auto;">
+    </br>
+    Unified Style and Subject-Driven Generation via Disentangled and Reward Learning
+</h3>
+
+<p align="center"> 
+<a href="https://bytedance.github.io/USO/"><img alt="Build" src="https://img.shields.io/badge/Project%20Page-USO-blue"></a> 
+<a href="https://arxiv.org/abs/2508.18966"><img alt="Build" src="https://img.shields.io/badge/Tech%20Report-USO-b31b1b.svg"></a>
+<a href="https://huggingface.co/bytedance-research/USO"><img src="https://img.shields.io/static/v1?label=%F0%9F%A4%97%20Hugging%20Face&message=Model&color=green"></a>
+<a href="https://huggingface.co/spaces/bytedance-research/USO"><img src="https://img.shields.io/static/v1?label=%F0%9F%A4%97%20Hugging%20Face&message=demo&color=orange"></a>
+</p>
+</p>
+
+><p align="center"> <span style="color:#137cf3; font-family: Gill Sans">Shaojin Wu,</span><sup></sup></a>  <span style="color:#137cf3; font-family: Gill Sans">Mengqi Huang,</span></a> <span style="color:#137cf3; font-family: Gill Sans">Yufeng Cheng,</span><sup></sup></a>  <span style="color:#137cf3; font-family: Gill Sans">Wenxu Wu,</span><sup></sup> </a> <span style="color:#137cf3; font-family: Gill Sans">Jiahe Tian,</span><sup></sup></a> <span style="color:#137cf3; font-family: Gill Sans">Yiming Luo,</span><sup></sup></a> <span style="color:#137cf3; font-family: Gill Sans">Fei Ding,</span></a> <span style="color:#137cf3; font-family: Gill Sans">Qian He</span></a> <br> 
+><span style="font-size: 13.5px">UXO Team</span><br> 
+><span style="font-size: 12px">Intelligent Creation Lab, Bytedance</span></p>
+
+### 🚩 Updates
+* **2025.09.03** 🎉 USO is now natively supported in ComfyUI, see official tutorial [USO in ComfyUI](https://docs.comfy.org/tutorials/flux/flux-1-uso) and our provided examples in `./workflow`. More tips are available in the [README below](https://github.com/bytedance/USO#%EF%B8%8F-comfyui-examples).
+<p align="center">
+<img src="assets/usoxcomfyui_official.jpeg" width=1024 height="auto">
+</p>
+
+* **2025.08.28** 🔥 The [demo](https://huggingface.co/spaces/bytedance-research/USO) of USO is released. Try it Now! ⚡️
+* **2025.08.28** 🔥 Update fp8 mode as a primary low vmemory usage support (please scroll down). Gift for consumer-grade GPU users. The peak Vmemory usage is ~16GB now.
+* **2025.08.27** 🔥 The [inference code](https://github.com/bytedance/USO) and [model](https://huggingface.co/bytedance-research/USO) of USO are released.
+* **2025.08.27** 🔥 The [project page](https://bytedance.github.io/USO) of USO is created.
+* **2025.08.27** 🔥 The [technical report](https://arxiv.org/abs/2508.18966) of USO is released.
+
+## 📖 Introduction
+Existing literature typically treats style-driven and subject-driven generation as two disjoint tasks: the former prioritizes stylistic similarity, whereas the latter insists on subject consistency, resulting in an apparent antagonism. We argue that both objectives can be unified under a single framework because they ultimately concern the disentanglement and re-composition of “content” and “style”, a long-standing theme in style-driven research. To this end, we present USO, a Unified framework for Style driven and subject-driven GeneratiOn. First, we construct a large-scale triplet dataset consisting of content images, style images, and their corresponding stylized content images. Second, we introduce a disentangled learning scheme that simultaneously aligns style features and disentangles content from style through two complementary objectives, style-alignment training and content–style disentanglement training. Third, we incorporate a style reward-learning paradigm to further enhance the model’s performance.
+<p align="center">
+    <img src="assets/teaser.webp" width="1024"/>
+</p>
+
+## ⚡️ Quick Start
+
+### 🔧 Requirements and Installation
+
+Install the requirements
+```bash
+## create a virtual environment with python >= 3.10 <= 3.12, like
+python -m venv uso_env
+source uso_env/bin/activate
+## or
+conda create -n uso_env python=3.10 -y
+conda activate uso_env
+
+## install torch
+## recommended version:
+pip install torch==2.4.0 torchvision==0.19.0 --index-url https://download.pytorch.org/whl/cu124 
+
+## then install the requirements by you need
+pip install -r requirements.txt # legacy installation command
+```
+
+Then download checkpoints:
+```bash
+# 1. set up .env file
+cp example.env .env
+
+# 2. set your huggingface token in .env (open the file and change this value to your token)
+HF_TOKEN=your_huggingface_token_here
+
+#3. download the necessary weights (comment any weights you don't need)
+pip install huggingface_hub
+python ./weights/downloader.py
+```
+- **IF YOU HAVE WEIGHTS, COMMENT OUT WHAT YOU DON'T NEED IN ./weights/downloader.py**
+
+### ✍️ Inference
+* Start from the examples below to explore and spark your creativity. ✨
+```bash
+# the first image is a content reference, and the rest are style references.
+
+# for subject-driven generation
+python inference.py --prompt "The man in flower shops carefully match bouquets, conveying beautiful emotions and blessings with flowers. " --image_paths "assets/gradio_examples/identity1.jpg" --width 1024 --height 1024
+# for style-driven generation
+# please keep the first image path empty
+python inference.py --prompt "A cat sleeping on a chair." --image_paths "" "assets/gradio_examples/style1.webp" --width 1024 --height 1024
+# for style-subject driven generation (or set the prompt to empty for layout-preserved generation)
+python inference.py --prompt "The woman gave an impassioned speech on the podium." --image_paths "assets/gradio_examples/identity2.webp" "assets/gradio_examples/style2.webp" --width 1024 --height 1024
+# for multi-style generation
+# please keep the first image path empty
+python inference.py --prompt "A handsome man." --image_paths "" "assets/gradio_examples/style3.webp" "assets/gradio_examples/style4.webp" --width 1024 --height 1024
+
+# for low vram:
+python inference.py --prompt "your propmt" --image_paths "your_image.jpg" --width 1024 --height 1024 --offload --model_type flux-dev-fp8 
+```
+* You can also compare your results with the results in the `assets/gradio_examples` folder.
+
+* For more examples, visit our [project page](https://bytedance.github.io/USO) or try the live [demo](https://huggingface.co/spaces/bytedance-research/USO).
+
+### 🌟 Gradio Demo
+
+```bash
+python app.py
+```
+
+**For low vmemory usage**, please pass the `--offload` and `--name flux-dev-fp8` args. The peak memory usage will be 16GB (Single reference) ~ 18GB (Multi references).
+
+```bash
+# please use FLUX_DEV_FP8 replace FLUX_DEV
+export FLUX_DEV_FP8="YOUR_FLUX_DEV_PATH"
+
+python app.py --offload --name flux-dev-fp8
+```
+
+## 🌈 More examples
+We provide some prompts and results to help you better understand the model. You can check our [paper](https://arxiv.org/abs/2508.18966) or [project page](https://bytedance.github.io/USO/) for more visualizations.
+
+#### Subject/Identity-driven generation
+<details>
+<summary>If you want to place a subject into new scene, please use natural language like "A dog/man/woman is doing...". If you only want to transfer the style but keep the layout, please an use instructive prompt like "Transform the style into ... style". For portraits-preserved generation, USO excels at producing high skin-detail images. A practical guideline: use half-body close-ups for half-body prompts, and full-body images when the pose or framing changes significantly. </summary>
+<p align="center">
+    <img src="assets/show_case1.webp" width="1024"/>
+<p>
+<p align="center">
+    <img src="assets/show_case2.webp" width="1024"/>
+</p>
+<p align="center">
+    <img src="assets/show_case3.webp" width="1024"/>
+</p>
+<p align="center">
+    <img src="assets/show_case4.webp" width="1024"/>
+</p>
+</details>
+
+
+#### Style-driven generation
+<details>
+<summary>Just upload one or two style images, and use natural language to create want you want. USO will generate images follow your prompt and match the style you uploaded. </summary>
+<p align="center">
+    <img src="assets/show_case5.webp" width="1024"/>
+<p>
+<p align="center">
+    <img src="assets/show_case6.webp" width="1024"/>
+</p>
+</details>
+
+#### Style-subject driven generation
+<details>
+<summary>USO can stylize a single content reference with one or two style refs. For layout-preserved generation, just set the prompt to empty. </summary>
+`Layout-preserved generation`
+<p align="center">
+    <img src="assets/show_case7.webp" width="1024"/>
+<p>
+
+`Layout-shifted generation`
+<p align="center">
+    <img src="assets/show_case8.webp" width="1024"/>
+</p>
+</details>
+
+## ⚙️ ComfyUI examples
+We’re pleased that USO now has native support in ComfyUI. For a quick start, please refer to the official tutorials [USO in ComfyUI](https://docs.comfy.org/tutorials/flux/flux-1-uso). To help you reproduce and match the results, we’ve provided several examples in `./workflows`, including **workflows** and their **inputs** and outputs, so you can quickly get familiar with what USO can do. With USO now fully compatible with the ComfyUI ecosystem, you can combine it with other plugins like ControlNet and LoRA. **We welcome community contributions of more workflows and examples.**
+
+Now you can easily run USO in ComfyUI. Just update ComfyUI to the latest version (0.3.57), and you’ll find USO in the official templates.
+<p align="center">
+    <img src="assets/comfyui_template.png" width=1024 height="auto">
+</p>
+
+More examples are provided below:
+<p align="center">
+<img src="assets/usoxcomfyui.webp" width=1024 height="auto">
+</p>
+
+**Identity preserved**
+<p align="center">
+    <img src="workflow/example1.png" width=1024 height="auto">
+</p>
+
+Download the image above and drag it into ComfyUI to load the corresponding [workflow](workflow/example1.json). Input images can be found in `./workflow`
+
+**Identity stylized**
+<p align="center">
+    <img src="workflow/example3.png" width=1024 height="auto">
+</p>
+
+Download the image above and drag it into ComfyUI to load the corresponding [workflow](workflow/example3.json). Input images can be found in `./workflow`
+
+**Identity + style reference**
+<p align="center">
+    <img src="workflow/example2.png" width=1024 height="auto">
+</p>
+
+Download the image above and drag it into ComfyUI to load the corresponding [workflow](workflow/example2.json). Input images can be found in `./workflow`
+
+**Single style reference**
+<p align="center">
+    <img src="workflow/example4.png" width=1024 height="auto">
+</p>
+
+Download the image above and drag it into ComfyUI to load the corresponding [workflow](workflow/example4.json). Input images can be found in `./workflow`
+<p align="center">
+    <img src="workflow/example6.png" width=1024 height="auto">
+</p>
+
+Download the image above and drag it into ComfyUI to load the corresponding [workflow](workflow/example6.json). Input images can be found in `./workflow`
+
+**Multiple style reference**
+<p align="center">
+    <img src="workflow/example5.png" width=1024 height="auto">
+</p>
+
+Download the image above and drag it into ComfyUI to load the corresponding [workflow](workflow/example5.json). Input images can be found in `./workflow`
+
+## 📄 Disclaimer
+<p>
+  We open-source this project for academic research. The vast majority of images 
+  used in this project are either generated or from open-source datasets. If you have any concerns, 
+  please contact us, and we will promptly remove any inappropriate content. 
+  Our project is released under the Apache 2.0 License. If you apply to other base models, 
+  please ensure that you comply with the original licensing terms. 
+  <br><br>This research aims to advance the field of generative AI. Users are free to 
+  create images using this tool, provided they comply with local laws and exercise 
+  responsible usage. The developers are not liable for any misuse of the tool by users.</p>
+
+## 🚀 Updates
+For the purpose of fostering research and the open-source community, we plan to open-source the entire project, encompassing training, inference, weights, dataset etc. Thank you for your patience and support! 🌟
+- [x] Release technical report.
+- [x] Release github repo.
+- [x] Release inference code.
+- [x] Release model checkpoints.
+- [x] Release huggingface space demo.
+- Release training code.
+- Release dataset.
+
+##  Citation
+If USO is helpful, please help to ⭐ the repo.
+
+If you find this project useful for your research, please consider citing our paper:
+```bibtex
+@article{wu2025uso,
+    title={USO: Unified Style and Subject-Driven Generation via Disentangled and Reward Learning},
+    author={Shaojin Wu and Mengqi Huang and Yufeng Cheng and Wenxu Wu and Jiahe Tian and Yiming Luo and Fei Ding and Qian He},
+    year={2025},
+    eprint={2508.18966},
+    archivePrefix={arXiv},
+    primaryClass={cs.CV},
+}
+```

app.py

@@ -0,0 +1,263 @@
+# Copyright (c) 2025 Bytedance Ltd. and/or its affiliates. All rights reserved.
+
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+
+#     http://www.apache.org/licenses/LICENSE-2.0
+
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import dataclasses
+import json
+import os
+from pathlib import Path
+
+from dotenv import load_dotenv
+load_dotenv() 
+
+
+import gradio as gr
+import torch
+
+from uso.flux.pipeline import USOPipeline
+from transformers import SiglipVisionModel, SiglipImageProcessor
+
+
+with open("assets/uso_text.svg", "r", encoding="utf-8") as svg_file:
+    text_content = svg_file.read()
+
+with open("assets/uso_logo.svg", "r", encoding="utf-8") as svg_file:
+    logo_content = svg_file.read()
+
+title = f"""
+<div style="display: flex; align-items: center; justify-content: center;">
+    <span style="transform: scale(0.7);margin-right: -5px;">{text_content}</span>    
+    <span style="font-size: 1.8em;margin-left: -10px;font-weight: bold; font-family: Gill Sans;">by UXO Team</span>
+    <span style="margin-left: 0px; transform: scale(0.85); display: inline-block;">{logo_content}</span>
+</div>
+""".strip()
+
+badges_text = r"""
+<div style="text-align: center; display: flex; justify-content: center; gap: 5px;">
+<a href="https://github.com/bytedance/USO"><img src="https://img.shields.io/static/v1?label=GitHub&message=Code&color=green&logo=github"></a>
+<a href="https://bytedance.github.io/USO/"><img alt="Build" src="https://img.shields.io/badge/Project%20Page-USO-yellow"></a>
+<a href="https://arxiv.org/abs/2504.02160"><img alt="Build" src="https://img.shields.io/badge/arXiv%20paper-USO-b31b1b.svg"></a>
+<a href="https://huggingface.co/bytedance-research/USO"><img src="https://img.shields.io/static/v1?label=%F0%9F%A4%97%20Hugging%20Face&message=Model&color=orange"></a>
+</div>
+""".strip()
+
+tips = """
+### What is USO and How to use?
+🎨 USO is a unified style-subject optimized customization model and the latest addition to the UXO family (<a href='https://github.com/bytedance/USO' target='_blank'> USO</a> and <a href='https://github.com/bytedance/UNO' target='_blank'> UNO</a>). 
+It can freely combine any subjects with any styles in any scenarios.
+
+💡 We provide step-by-step instructions in our <a href='https://github.com/bytedance/USO#-more-examples' target='_blank'> Github Repo</a>.
+Additionally, try the examples provided below the demo to quickly get familiar with USO and inspire your creativity!
+
+### Updates
+🔥 **2025.09.04** USO now has native support in ComfyUI (see <a href='https://docs.comfy.org/tutorials/flux/flux-1-uso' target='_blank'>ComfyUI's official documentation</a> for details). For more information, please also check out our <a href='https://github.com/bytedance/USO?tab=readme-ov-file#%EF%B8%8F-comfyui-examples' target='_blank'>GitHub Repo</a>.
+
+<details>
+<summary style="cursor: pointer; color: #d34c0e; font-weight: 500;">The model is trained on 1024x1024 resolution and supports 3 types of usage. Tips:</summary>
+
+* **Only content img**: support following types:
+  * Subject/Identity-driven (supports natural prompt, e.g., *A clock on the table.* *The woman near the sea.*, excels in producing **photorealistic portraits**)
+  * Style edit (layout-preserved): *Transform the image into Ghibli style/Pixel style/Retro comic style/Watercolor painting style...*.
+  * Style edit (layout-shift): *Ghibli style, the man on the beach.*.
+* **Only style img**: Reference input style and generate anything following prompt. Excelling in this and further support multiple style references (in beta).
+* **Content img + style img**: Place the content into the desired style. 
+  * Layout-preserved: set prompt to **empty**.
+  * Layout-shift: using natural prompt.</details>"""
+
+star = """
+### If USO is helpful, please help to ⭐ our <a href='https://github.com/bytedance/USO' target='_blank'> Github Repo</a>. Thanks a lot!"""
+
+def get_examples(examples_dir: str = "assets/examples") -> list:
+    examples = Path(examples_dir)
+    ans = []  
+    for example in examples.iterdir():
+        if not example.is_dir() or len(os.listdir(example)) == 0:
+            continue
+        with open(example / "config.json") as f:
+            example_dict = json.load(f)
+
+
+        example_list = []
+        example_list.append(example_dict["prompt"])  # prompt
+
+        for key in ["image_ref1", "image_ref2", "image_ref3"]:
+            if key in example_dict:
+                example_list.append(str(example / example_dict[key]))
+            else:
+                example_list.append(None)
+
+        example_list.append(example_dict["seed"])
+        ans.append(example_list)
+    return ans
+
+
+def create_demo(
+    model_type: str,
+    device: str = "cuda" if torch.cuda.is_available() else ("mps" if torch.backends.mps.is_available() else "cpu"),
+    offload: bool = False,
+):
+
+    # hf_download set to false to prevent download of weights
+    pipeline = USOPipeline(
+        model_type, device, offload, only_lora=True, lora_rank=128, hf_download=False
+    )
+    print("USOPipeline loaded successfully")
+
+
+
+
+    # ⚠️ Weights now load from local paths via .env instead of downloading
+    siglip_path = os.getenv("SIGLIP_PATH", "google/siglip-so400m-patch14-384")
+    siglip_processor = SiglipImageProcessor.from_pretrained(siglip_path)
+    siglip_model = SiglipVisionModel.from_pretrained(siglip_path)
+    
+    
+    siglip_model.eval()
+    siglip_model.to(device)
+    pipeline.model.vision_encoder = siglip_model
+    pipeline.model.vision_encoder_processor = siglip_processor
+    print("SigLIP model loaded successfully")
+
+    with gr.Blocks() as demo:
+        gr.Markdown(title)
+        gr.Markdown(badges_text)
+        gr.Markdown(tips)
+        with gr.Row():
+            with gr.Column():
+                prompt = gr.Textbox(label="Prompt", value="A beautiful woman.")
+                with gr.Row():
+                    image_prompt1 = gr.Image(
+                        label="Content Reference Img", visible=True, interactive=True, type="pil"
+                    )
+                    image_prompt2 = gr.Image(
+                        label="Style Reference Img", visible=True, interactive=True, type="pil"
+                    )
+                    image_prompt3 = gr.Image(
+                        label="Extra Style Reference Img (Beta)", visible=True, interactive=True, type="pil"
+                    )
+
+                with gr.Row():
+                    with gr.Row():
+                        width = gr.Slider(
+                            512, 1536, 1024, step=16, label="Generation Width"
+                        )
+                        height = gr.Slider(
+                            512, 1536, 1024, step=16, label="Generation Height"
+                        )
+                with gr.Row():
+                    with gr.Row():
+                        keep_size = gr.Checkbox(
+                            label="Keep input size",
+                            value=False,
+                            interactive=True
+                        )
+                    with gr.Column():
+                        gr.Markdown("Set it to True if you only need style editing or want to keep the layout.")
+
+                with gr.Accordion("Advanced Options", open=True):
+                    with gr.Row():
+                        num_steps = gr.Slider(
+                            1, 50, 25, step=1, label="Number of steps"
+                        )
+                        guidance = gr.Slider(
+                            1.0, 5.0, 4.0, step=0.1, label="Guidance", interactive=True
+                        )
+                        content_long_size = gr.Slider(
+                            0, 1024, 512, step=16, label="Content reference size"
+                        )                        
+                        seed = gr.Number(-1, label="Seed (-1 for random)")
+
+                generate_btn = gr.Button("Generate")
+                gr.Markdown(star)
+
+            with gr.Column():
+                output_image = gr.Image(label="Generated Image")
+                download_btn = gr.File(
+                    label="Download full-resolution", type="filepath", interactive=False
+                )
+                gr.Markdown(
+                    """
+                    ### ❗️ Important Usage Tips:
+                    - **Input Prompt**: Unless you only need Style Editing ("Transform the style into..."), use natural language ("A dog/man/woman is doing...") instead of instruction descriptions of subject, identity, or style.
+                    - **Input Content Image**: For portrait-preserving generation, USO excels at producing images with high skin detail. A practical guideline: use half-body close-ups when your prompt specifies a half-body subject, and full-body images—especially when the pose changes significantly.
+                    """
+                )                
+
+            inputs = [
+                prompt,
+                image_prompt1,
+                image_prompt2,
+                image_prompt3,
+                seed,                     
+                width,
+                height,
+                guidance,
+                num_steps,
+                keep_size,
+                content_long_size,
+            ]
+            generate_btn.click(
+                fn=pipeline.gradio_generate,
+                inputs=inputs,
+                outputs=[output_image, download_btn],
+            )   
+
+        # example_text = gr.Text("", visible=False, label="Case For:")
+        examples = get_examples("./assets/gradio_examples")
+
+        gr.Examples(
+            examples=examples,
+            inputs=[
+                prompt,
+                image_prompt1,
+                image_prompt2,
+                image_prompt3,
+                seed,
+            ],
+            # cache_examples='lazy',
+            outputs=[output_image, download_btn],
+            fn=pipeline.gradio_generate,
+            label='row 1-4: identity/subject-driven; row 5-7: style-subject-driven; row 8-9: style-driven; row 10-12: multi-style-driven task; row 13: txt2img',
+            examples_per_page=15
+        )
+
+        with gr.Accordion("Local Gradio Demo for Developers", open=False):
+            gr.Markdown(
+                'Please refer to our GitHub repository to [run the USO gradio demo locally](https://github.com/bytedance/USO?tab=readme-ov-file#-gradio-demo).'
+            )
+    return demo
+
+
+if __name__ == "__main__":
+    from typing import Literal
+
+    from transformers import HfArgumentParser
+
+    @dataclasses.dataclass
+    class AppArgs:
+        name: Literal["flux-dev", "flux-dev-fp8", "flux-schnell", "flux-krea-dev"] = "flux-dev"
+        device: Literal["cuda", "cpu", "mps"] = "cuda" if torch.cuda.is_available() else ("mps" if torch.backends.mps.is_available() else "cpu")
+        offload: bool = dataclasses.field(
+            default=False,
+            metadata={
+                "help": "If True, sequantial offload the models(ae, dit, text encoder) to CPU if not used."
+            },
+        )
+        port: int = 7860
+
+    parser = HfArgumentParser([AppArgs])
+    args_tuple = parser.parse_args_into_dataclasses()  # type: tuple[AppArgs]
+    args = args_tuple[0]
+
+    demo = create_demo(args.name, args.device, args.offload)
+    demo.launch(server_port=args.port)

assets/gradio_examples/1identity/config.json

@@ -0,0 +1,6 @@
+{
+    "prompt": "The girl is riding a bike in a street.",
+    "seed": 3407,
+    "usage": "Identity-driven",
+    "image_ref1": "./ref.webp"
+}

assets/gradio_examples/2identity/config.json

@@ -0,0 +1,6 @@
+{
+    "prompt": "The man in flower shops carefully match bouquets, conveying beautiful emotions and blessings with flowers.",
+    "seed": 3407,
+    "usage": "Identity-driven",
+    "image_ref1": "./ref.jpg"
+}

assets/gradio_examples/3identity/config.json

@@ -0,0 +1,6 @@
+{
+    "prompt": "Transform the image into Ghibli style.",
+    "seed": 3407,
+    "usage": "Identity-driven",
+    "image_ref1": "./ref.webp"
+}

assets/gradio_examples/4subject/config.json

@@ -0,0 +1,6 @@
+{
+    "prompt": "Wool felt style, a clock in the jungle.",
+    "seed": 3407,
+    "usage": "Subject-driven",
+    "image_ref1": "./ref.jpg"
+}

assets/gradio_examples/5style_subject/config.json

@@ -0,0 +1,7 @@
+{
+    "prompt": "",
+    "seed": 321,
+    "usage": "Style-subject-driven (layout-preserved)",
+    "image_ref1": "./ref1.webp",
+    "image_ref2": "./ref2.webp"
+}

assets/gradio_examples/6style_subject/config.json

@@ -0,0 +1,7 @@
+{
+    "prompt": "The woman gave an impassioned speech on the podium.",
+    "seed": 321,
+    "usage": "Style-subject-driven (layout-shifted)",
+    "image_ref1": "./ref1.webp",
+    "image_ref2": "./ref2.webp"
+}

assets/gradio_examples/7style_subject/config.json

@@ -0,0 +1,7 @@
+{
+    "prompt": "The man gave an impassioned speech on the podium. ",
+    "seed": 42,
+    "usage": "Style-subject-driven (layout-shifted)",
+    "image_ref1": "./ref1.webp",
+    "image_ref2": "./ref2.webp"
+}

assets/gradio_examples/8style/config.json

@@ -0,0 +1,6 @@
+{
+    "prompt": "A cat sleeping on a chair.",
+    "seed": 3407,
+    "usage": "Style-driven",
+    "image_ref2": "./ref.webp"
+}

assets/gradio_examples/9style/config.json

@@ -0,0 +1,6 @@
+{
+    "prompt": "A beautiful woman.",
+    "seed": 3407,
+    "usage": "Style-driven",
+    "image_ref2": "./ref.webp"
+}

assets/gradio_examples/z1_mix_style/config.json

@@ -0,0 +1,7 @@
+{
+    "prompt": "A man.",
+    "seed": 321,
+    "usage": "Multi-style-driven",
+    "image_ref2": "./ref1.webp",
+    "image_ref3": "./ref2.webp"
+}

assets/gradio_examples/z2_mix_style/config.json

@@ -0,0 +1,7 @@
+{
+    "prompt": "Boat on water.",
+    "seed": 321,
+    "usage": "Multi-style-driven",
+    "image_ref2": "./ref1.png",
+    "image_ref3": "./ref2.png"
+}

assets/gradio_examples/z3_mix_style/config.json

@@ -0,0 +1,8 @@
+{
+    "prompt": "",
+    "seed": 321,
+    "usage": "Multi-style-driven",
+    "image_ref1": "./ref1.jpg",
+    "image_ref2": "./ref2.png",
+    "image_ref3": "./ref3.png"
+}

assets/gradio_examples/z4_t2i/config.json

@@ -0,0 +1,5 @@
+{
+    "prompt": "A beautiful woman.",
+    "seed": 0,
+    "usage": "Text-to-image"
+}