The installation guide is inherited from StreamDiffusion.
The initial release only tested the Linux env with conda/Docker enviroment.
git clone https://github.com/Jeff-LiangF/streamv2v.git
You can install StreamV2V via conda, or Docker(explanation below).
# Using conda (Recommended)
conda create -n streamv2v python=3.10
conda activate streamv2v
Select the appropriate version for your system. Check the Pytorch doc.
# CUDA 11.8
pip3 install torch==2.1.0 torchvision==0.16.0 xformers --index-url https://download.pytorch.org/whl/cu118
# CUDA 12.1
pip3 install torch==2.1.0 torchvision==0.16.0 xformers --index-url https://download.pytorch.org/whl/cu121
Install other dependencies
pip install -r requirements.txt
python setup.py develop easy_install streamv2v[tensorrt]
python -m streamv2v.tools.install-tensorrt
# for Latest Version (recommended)
pip install git+https://github.com/Jeff-LiangF/streamv2v.git@main#egg=streamv2v[tensorrt]
# Install TensorRT extension
python -m streamv2v.tools.install-tensorrt
(Only for Windows) You may need to install pywin32.
pip install --force-reinstall pywin32
git clone https://github.com/Jeff-LiangF/streamv2v.git
cd streamv2v
docker build -t streamv2v:latest -f Dockerfile .
docker run --gpus all -it -v $(pwd):/home/ubuntu/streamv2v streamv2v:latest