bytedance/UNO
ByteDance's UNO is a diffusion model enabling universal single and multi-subject image customization via in-context learning.
Not currently ranked — collecting fresh signals.
star history
UNO implements a less-to-more generalization approach for subject-driven image generation, allowing flexible conditioning on both single and multiple subjects. Built on diffusion transformers like FLUX, it enables customizable text-to-image synthesis with high identity consistency across subjects. The project includes model weights, a 1M sample dataset, and a Hugging Face demo space.
Frequently asked
- What is bytedance/UNO?
- ByteDance's UNO is a diffusion model enabling universal single and multi-subject image customization via in-context learning.
- Is UNO open source?
- Yes — bytedance/UNO is open source, released under the Apache-2.0 license.
- What language is UNO written in?
- bytedance/UNO is primarily written in Python.
- How popular is UNO?
- bytedance/UNO has 1.4k stars on GitHub.
- Where can I find UNO?
- bytedance/UNO is on GitHub at https://github.com/bytedance/UNO.