NVIDIA RTX Video Super Resolution for ComfyUI, optimized for long videos with disk-backed processing and low RAM usage. Supports AI upscaling, denoising, deblurring, H.264/HEVC NVENC encoding, audio preservation, and an integrated video preview.
Python
51
13 commits
updated Oct 12, 2026
NVIDIA RTX Video Super Resolution for ComfyUI, optimized for long videos with low RAM usage.
Uses NVIDIA's nvidia-vfx SDK to upscale, denoise, and deblur videos without loading the entire video into memory.
Clone this repository into ComfyUI/custom_nodes/:
git clone https://github.com/tritant/ComfyUI_NVIDIA_RTX_VSR_Long_Video.git
Install the dependencies using ComfyUI's Python environment:
python -m pip install -r requirements.txt
Requirements: NVIDIA RTX GPU, compatible CUDA/PyTorch environment, nvidia-vfx >= 0.2, and FFmpeg.
For NVENC encoding, FFmpeg must support h264_nvenc or hevc_nvenc.
Connect a native Load Video node or any compatible VIDEO output to RTX VSR Long Video.
Load Video
↓
RTX VSR Long Video
↓
VIDEO output → Other video nodes
Select your upscale factor, NVIDIA quality mode, strength, and optional Deblur/Denoise preprocessing.
The finished MP4 is automatically saved to:
ComfyUI/output/NVIDIA_RTX_VSR_Long_Video/
The node also provides an integrated preview, a file path output, and a native VIDEO output for further processing.
Instead of keeping thousands of uncompressed frames in RAM, the node:
Memory usage stays manageable even with long, high-resolution videos.
Powered by NVIDIA Video Effects SDK.
NVIDIA RTX VSR technology and models belong to NVIDIA. This project is an independent ComfyUI integration.
NVIDIA RTX Video Super Resolution for ComfyUI, optimized for long videos with disk-backed processing and low RAM usage. Supports AI upscaling, denoising, deblurring, H.264/HEVC NVENC encoding, audio preservation, and an integrated video preview.
Python
51
13 commits
updated Oct 12, 2026
NVIDIA RTX Video Super Resolution for ComfyUI, optimized for long videos with low RAM usage.
Uses NVIDIA's nvidia-vfx SDK to upscale, denoise, and deblur videos without loading the entire video into memory.
Clone this repository into ComfyUI/custom_nodes/:
git clone https://github.com/tritant/ComfyUI_NVIDIA_RTX_VSR_Long_Video.git
Install the dependencies using ComfyUI's Python environment:
python -m pip install -r requirements.txt
Requirements: NVIDIA RTX GPU, compatible CUDA/PyTorch environment, nvidia-vfx >= 0.2, and FFmpeg.
For NVENC encoding, FFmpeg must support h264_nvenc or hevc_nvenc.
Connect a native Load Video node or any compatible VIDEO output to RTX VSR Long Video.
Load Video
↓
RTX VSR Long Video
↓
VIDEO output → Other video nodes
Select your upscale factor, NVIDIA quality mode, strength, and optional Deblur/Denoise preprocessing.
The finished MP4 is automatically saved to:
ComfyUI/output/NVIDIA_RTX_VSR_Long_Video/
The node also provides an integrated preview, a file path output, and a native VIDEO output for further processing.
Instead of keeping thousands of uncompressed frames in RAM, the node:
Memory usage stays manageable even with long, high-resolution videos.
Powered by NVIDIA Video Effects SDK.
NVIDIA RTX VSR technology and models belong to NVIDIA. This project is an independent ComfyUI integration.