FacesImage generation

PuLID

Preserves a person's face when generating images from one photo, with less damage to style and background. Versions exist for SDXL and FLUX; the latter runs on a 16 GB card.

Developer
ByteDance, China
First release
May 2024
Latest release
Oct 2024
Sizes
adapters for SDXL and FLUX.1-dev
License
Commercial use with conditionsWeights are Apache 2.0, but the FLUX version is subject to the non-commercial FLUX.1-dev license, and the InsightFace face recognizer is for research only
Running
On your own serverNeeds a GPU
Industries
Marketing and content, Media and production
Deploy on your server Hugging Face GitHub

What it does

  • Portraits and avatars from one photo
  • Ad characters with a recognizable face
  • Photoshoot prototypes

Where it is used

Marketing and creativeDesign studios

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions
no versions
1 GPUOne GPU with 16–80 GB — mid-size versions
fits
ClusterServer with several GPUs — flagship versions
no versions

Versions

  1. PuLID v1.1 и PuLID-FLUX v0.9.1
  2. PuLID-FLUX v0.9.0
  3. PuLID v1 для SDXL

How to run it

On your own serverWeights are downloaded from Hugging Face and served with vLLM (load and API) or llama.cpp (modest hardware). The model then runs in a closed network with no per-request fees.Hugging Face
How much hardware you needCalculate the VRAM for the model size, context length and number of concurrent requests.Open the hardware calculator

I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems.

Frequently asked questions

Can PuLID be used in a commercial project?

With conditions. License: Weights are Apache 2.0, but the FLUX version is subject to the non-commercial FLUX.1-dev license, and the InsightFace face recognizer is for research only. Restrictions vary — region, company revenue, attribution requirements. Have a lawyer check the terms before a commercial launch.

What hardware does PuLID need?

At minimum: One GPU with 16–80 GB — mid-size versions. Without a GPU the model is not practical. You can calculate the exact VRAM for your model size and context in the hardware calculator.

Does PuLID support Russian?

Language does not matter for this model: it does not work with text.

Where can I download PuLID and what does it cost?

The PuLID weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.

How I deploy it for clients

  1. SelectionI pick the model size for your task and hardware and test it on your examples.
  2. DeploymentI deploy it on your server or in a closed network and provide an API.
  3. Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
  4. IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.

Similar models

FacesInstantIDInstantX (Xiaohongshu) · ChinaNon-commercial only

Generates images with a specific person's face from a single photo, without fine-tuning. Popular in ComfyUI, but the weights are for research only.

DetailsFacesInfiniteYouByteDance · ChinaNon-commercial only

FLUX-based image generation that preserves a face: follows the prompt better and less often pastes the face like a sticker. Research-only license.

DetailsFacesIP-Adapter-FaceIDTencent AI Lab (h94) · ChinaNon-commercial only

One of the first adapters that transfer a face from a photo into a generated image. The SD 1.5 versions run on low-end cards, but the weights are non-commercial.

Details

Source: huggingface.co/guozinan/PuLID. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.