Stable Video Diffusion 1.1 TensorRT
This repository hosts the TensorRT version of the Stable Video Diffusion (SVD) 1.1 Image-to-Video model.
Model Details
Please see Stable Video Diffusion (SVD) 1.1 Image-to-Video for the full model details.
This model is intended for research purposes only and should not be used in any way that violates Stability AI's Acceptable Use Policy.
Performance
SVD-XT 1.1 (25 frames, 25 steps)
| A100 80GB PCI | A100 80GB SXM | H100 80GB PCI | |
|---|---|---|---|
| VAE Encoder | 66.70 ms | 65.68 ms | 49.07 ms |
| CLIP | 105.41 ms | 53.20 ms | 91.32 ms |
| UNet x 25 | 30,367.73 ms | 27,489.88 ms | 19,102.98 ms |
| VAE Decoder | 4,663.63 ms | 4,544.12 ms | 3,382.62 ms |
| Total E2E | 35,258.38 ms | 32,166.41 ms | 22,644.73 ms |
Usage Example
- Clone TensorRT and this repo then launch NGC container
git clone https://github.com/rajeevsrao/TensorRT.git
cd TensorRT
git checkout release/svd
git lfs install
git clone https://huggingface.co./stabilityai/stable-video-diffusion-img2vid-xt-1-1-tensorrt
docker run --rm -it --gpus all -v $PWD:/workspace nvcr.io/nvidia/pytorch:23.12-py3 /bin/bash
- Install libraries and requirements
cd demo/Diffusion
python3 -m pip install --upgrade pip
pip3 install -r requirements.txt
python3 -m pip install --pre --upgrade --extra-index-url https://pypi.nvidia.com tensorrt
- Authenticate with huggingface
huggingface-cli login
- Perform TensorRT optimized inference:
python3 demo_img2vid.py \
--version svd-xt-1.1 \
--onnx-dir /workspace/stable-video-diffusion-img2vid-xt-1-1-tensorrt \
--engine-dir engine-svd-xt-1-1 \
--build-static-batch \
--use-cuda-graph \
--input-image https://www.hdcarwallpapers.com/walls/2018_chevrolet_camaro_zl1_nascar_race_car_2-HD.jpg