Skip to content

Latest commit

 

History

History
72 lines (46 loc) · 3.01 KB

readme.md

File metadata and controls

72 lines (46 loc) · 3.01 KB
Raw

ComfyUI Rife TensorRT ⚡

python cuda trt by-nc-sa/4.0

node

This project provides a TensorRT implementation of RIFE for ultra fast frame interpolation inside ComfyUI

This project is licensed under CC BY-NC-SA, everyone is FREE to access, use, modify and redistribute with the same license.

If you like the project, please give me a star! ⭐

⏱️ Performance

Note: The following results were benchmarked on FP16 engines inside ComfyUI, using 2000 frames consisting of 2 alternating similar frames, averaged 2-3 times

Device Rife Engine Resolution Multiplier FPS
H100 rife49_ensemble_True_scale_1_sim 512 x 512 2 45
H100 rife49_ensemble_True_scale_1_sim 512 x 512 4 57
H100 rife49_ensemble_True_scale_1_sim 1280 x 1280 2 21

🚀 Installation

Navigate to the ComfyUI /custom_nodes directory

git clone https://github.com/yuvraj108c/ComfyUI-Rife-Tensorrt
cd ./ComfyUI-Rife-Tensorrt
pip install -r requirements.txt

🛠️ Building Tensorrt Engine

  1. Download one of the following onnx models:

  2. Edit onnx/trt paths inside export_trt.py and build tensorrt engine by running:

    • python export_trt.py

  3. Place the exported engine inside ComfyUI /models/tensorrt/rife directory

☀️ Usage

  • Insert node by Right Click -> tensorrt -> Rife Tensorrt
  • Image resolutions between 256x256 and 3840x3840 will work with the tensorrt engines

🤖 Environment tested

  • Ubuntu 22.04 LTS, Cuda 12.4, Tensorrt 10.4.0, Python 3.10, RTX 3070 GPU
  • Windows (Not tested, but should work)

👏 Credits

License

Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0)