Hugging Face's logo

Skywork
/
Unipic3-Consistency-Model

Any-to-Any
Transformers
Diffusers
text-to-image
image-editing
image-understanding
vision-language
multimodal
unified-model
Model card Files
xet
 Community
1
OrlandoHugBot commited on
Commit
4e9d898
·
verified ·
1 Parent(s): ce4e34c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -0
README.md CHANGED
@@ -30,6 +30,7 @@ license: mit
30
 
31
  ## 📖 Introduction
32
  <div align="center"> <img src="unipic3.png" alt="Model Teaser" width="720"> </div>
 
33
  **UniPic3-Consistency-Model** is a few-step image editing and multi-image composition model based on **Consistency Flow Matching (CM)**.
34
  The model learns a *trajectory-consistent* mapping from noisy latent states to clean images, enabling stable generation with strong structural consistency.
35
  It is distilled from **UniPic-3** to support **fast inference (≤8 steps)** while preserving composition correctness.The model is especially suitable for scenarios requiring **geometric alignment** and **semantic coherence**, such as multi-image composition and human–object interaction (HOI).
 
30
 
31
  ## 📖 Introduction
32
  <div align="center"> <img src="unipic3.png" alt="Model Teaser" width="720"> </div>
33
+
34
  **UniPic3-Consistency-Model** is a few-step image editing and multi-image composition model based on **Consistency Flow Matching (CM)**.
35
  The model learns a *trajectory-consistent* mapping from noisy latent states to clean images, enabling stable generation with strong structural consistency.
36
  It is distilled from **UniPic-3** to support **fast inference (≤8 steps)** while preserving composition correctness.The model is especially suitable for scenarios requiring **geometric alignment** and **semantic coherence**, such as multi-image composition and human–object interaction (HOI).