• If you find PersonaLive useful or interesting, please give us a Star🌟! Your support drives us to keep improving.
• Fix bugs (If you encounter any issues, please feel free to open an issue or contact me! 🙏)
• Release training code.
• [2026年02月21日] 🥳 PersonaLive is accepted by CVPR2026 🎉.
• [2025年12月29日] 🔥 Enhance WebUI (Support reference image replacement).
• [2025年12月22日] 🔥 Supported streaming strategy in offline inference to generate long videos on 12GB VRAM!
• [2025年12月17日] 🔥 ComfyUI-PersonaLive is now supported! (Thanks to @okdalto)
• [2025年12月15日] 🔥 Release paper!
• [2025年12月12日] 🔥 Release inference code, config, and pretrained weights!

• This project is released for academic research only.
• Users must not use this repository to generate harmful, defamatory, or illegal content.
• The authors bear no responsibility for any misuse or legal consequences arising from the use of this tool.
• By using this code, you agree that you are solely responsible for any content generated.

Image 1

We present PersonaLive, a real-time and streamable diffusion framework capable of generating infinite-length portrait animations.

# clone this repo
git clone https://github.com/GVCLab/PersonaLive
cd PersonaLive
# Create conda environment
conda create -n personalive python=3.10
conda activate personalive
# Install packages with pip
pip install -r requirements_base.txt

Option 1: Download pre-trained weights of base models and other components (sd-image-variations-diffusers and sd-vae-ft-mse). You can run the following command to download weights automatically:

python tools/download_weights.py

Option 2: Download pre-trained weights into the ./pretrained_weights folder from one of the below URLs:

Finally, these weights should be organized as follows:

pretrained_weights
├── onnx
│ ├── unet_opt
│ │ ├── unet_opt.onnx
│ │ └── unet_opt.onnx.data
│ └── unet
├── personalive
│ ├── denoising_unet.pth
│ ├── motion_encoder.pth
│ ├── motion_extractor.pth
│ ├── pose_guider.pth
│ ├── reference_unet.pth
│ └── temporal_module.pth
├── sd-vae-ft-mse
│ ├── diffusion_pytorch_model.bin
│ └── config.json
├── sd-image-variations-diffusers
│ ├── image_encoder
│ │ ├── pytorch_model.bin
│ │ └── config.json
│ ├── unet
│ │ ├── diffusion_pytorch_model.bin
│ │ └── config.json
│ └── model_index.json
└── tensorrt
 └── unet_work.engine

Run offline inference with the default configuration:

python inference_offline.py

• -L: Max number of frames to generate. (Default: 100)
• --use_xformers: Enable xFormers memory efficient attention. (Default: True)
• --stream_gen: Enable streaming generation strategy. (Default: True)
• --reference_image: Path to a specific reference image. Overrides settings in config.
• --driving_video: Path to a specific driving video. Overrides settings in config.

⚠️ Note for RTX 50-Series (Blackwell) Users: xformers is not yet fully compatible with the new architecture. To avoid crashes, please disable it by running:

python inference_offline.py --use_xformers False

# install Node.js 18+
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.39.1/install.sh | bash
nvm install 18
source web_start.sh

Converting the model to TensorRT can significantly speed up inference (~ 2x ⚡️). Building the engine may take about 20 minutes depending on your device. Note that TensorRT optimizations may lead to slight variations or a small drop in output quality.

# Install packages with pip
pip install -r requirements_trt.txt
# Converting the model to TensorRT
python torch2trt.py

💡 PyCUDA Installation Issues: If you encounter a "Failed to build wheel for pycuda" error during the installation above, please follow these steps:

# Install PyCUDA manually using Conda (avoids compilation issues):
conda install -c conda-forge pycuda "numpy<2.0"
# Open requirements_trt.txt and comment out or remove the line "pycuda==2024年1月2日"
# Install other packages with pip
pip install -r requirements_trt.txt
# Converting the model to TensorRT
python torch2trt.py

⚠️ The provided TensorRT model is from an H100. We recommend ALL users (including H100 users) re-run python torch2trt.py locally to ensure best compatibility.

python inference_online.py --acceleration none (for RTX 50-Series) or xformers or tensorrt

Then open http://0.0.0.0:7860 in your browser. (*If http://0.0.0.0:7860 does not work well, try http://localhost:7860)

How to use: Upload Image ➡️ Fuse Reference ➡️ Start Animation ➡️ Enjoy! 🎉

Regarding Latency: Latency varies depending on your device's computing power. You can try the following methods to optimize it:

1. Lower the "Driving FPS" setting in the WebUI to reduce the computational workload.
2. You can increase the multiplier (e.g., set to num_frames_needed * 4 or higher) to better match your device's inference speed. https://github.com/GVCLab/PersonaLive/blob/6953d1a8b409f360a3ee1d7325093622b29f1e22/webcam/util.py#L73

Special thanks to the community for providing helpful setups! 🥂

• Windows + RTX 50-Series Guide: Thanks to @dknos for providing a detailed guide on running this project on Windows with Blackwell GPUs.

• TensorRT on Windows: If you are trying to convert TensorRT models on Windows, this discussion might be helpful. Special thanks to @MaraScott and @Jeremy8776 for their insights.

• ComfyUI: Thanks to @okdalto for helping implement the ComfyUI-PersonaLive support.

• Useful Scripts: Thanks to @suruoxi for implementing download_weights.py, and to @andchir for adding audio merging functionality.

If you find PersonaLive useful for your research, welcome to cite our work using the following BibTeX:

@article{li2025personalive,
 title={PersonaLive! Expressive Portrait Image Animation for Live Streaming},
 author={Li, Zhiyuan and Pun, Chi-Man and Fang, Chen and Wang, Jue and Cun, Xiaodong},
 journal={arXiv preprint arXiv:2512.11253},
 year={2025}
}

This code is mainly built upon Moore-AnimateAnyone, X-NeMo, StreamDiffusion, RAIN and LivePortrait, thanks to their invaluable contributions.