PuLID: Pure and Lightning ID Customization via Contrastive Alignment

Zinan Guo, Yanze Wu, Zhuowei Chen, Lang Chen, Qian He
2/3/2026

No official code found. Best implementation: ToTheBeginning/PuLID (3,550 stars, Apache-2.0, last commit 2025-07-31).

Abstract

We propose Pure and Lightning ID customization (PuLID), a novel tuning-free ID customization method for text-to-image generation. By incorporating a Lightning T2I branch with a standard diffusion one, PuLID introduces both contrastive alignment loss and accurate ID loss, minimizing disruption to the original model and ensuring high ID fidelity. Experiments show that PuLID achieves superior performance in both ID fidelity and editability. Another attractive property of PuLID is that the image elements (e.g., background, lighting, composition, and style) before and after the ID insertion are kept as consistent as possible. Codes and models are available at https://github.com/ToTheBeginning/PuLID

View on arXivView PDF

Code Implementations(1)

ToTheBeginning/PuLIDCommunity100%
Apache-2.0

[NeurIPS 2024] Official code for PuLID: Pure and Lightning ID Customization via Contrastive Alignment

3,550268PythonApr 17, 20241 years agoApache-2.0

Similar Papers

Tuning-Free Image Customization with Image and Text Guidance

87% match1 repo1482026

Inv-Adapter: ID Customization Generation via Image Inversion and Lightweight Adapter

85% match2024

TiNO-Edit: Timestep and Noise Optimization for Robust Diffusion-Based Image Editing

85% match1 repo442026

EditID: Training-Free Editable ID Customization for Text-to-Image Generation

85% match1 repo22025

AlignIT: Enhancing Prompt Alignment in Customization of Text-to-Image Models

84% match2024

Customization Assistant for Text-to-image Generation

84% match2026

Cite this paper

@article{guo2026pulid,
  title  = {PuLID: Pure and Lightning ID Customization via Contrastive Alignment},
  author = {Zinan Guo and Yanze Wu and Zhuowei Chen and Lang Chen and Qian He},
  year   = {2026},
  eprint = {2404.16022},
  archivePrefix = {arXiv},
  doi    = {10.48550/arXiv.2404.16022},
  url    = {https://arxiv.org/abs/2404.16022},
  journal = {NEURIPS 2024 2024}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2404.16022.svg)](https://codeofpaper.com/paper/2404.16022)

Discussion