ComfyUI > Nodes > ComfyUI-FlashVSR

ComfyUI Extension: ComfyUI-FlashVSR

Repo Name

ComfyUI-FlashVSR

Author
ailab (Account age: 999 days)
Nodes
View all nodes(2)
Latest Updated
2025-11-17
Github Stars
0.07K

How to Install ComfyUI-FlashVSR

Install this extension via the ComfyUI Manager by searching for ComfyUI-FlashVSR
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter ComfyUI-FlashVSR in the search bar
After installation, click the Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • 16GB VRAM to 80GB VRAM GPU machines
  • 400+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 200+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

ComfyUI-FlashVSR Description

ComfyUI-FlashVSR is a robust ComfyUI custom node leveraging the FlashVSR model to enable real-time diffusion-based video super-resolution, enhancing streaming applications.

ComfyUI-FlashVSR Introduction

ComfyUI-FlashVSR is an innovative extension designed to enhance video quality through real-time super-resolution processing. Built on the powerful FlashVSR model, this extension allows you to upscale video resolution effectively, making it particularly useful for streaming applications. By integrating with ComfyUI, it provides a seamless experience for AI artists looking to improve video clarity and detail without the need for extensive technical knowledge.

How ComfyUI-FlashVSR Works

At its core, ComfyUI-FlashVSR leverages the FlashVSR model, which is a diffusion-based framework for video super-resolution. This means it uses advanced algorithms to predict and enhance the details in video frames, effectively increasing their resolution. Imagine taking a low-resolution video and transforming it into a high-definition version, much like how a magnifying glass can reveal finer details in a picture. The extension processes video frames in real-time, ensuring that the enhanced video is ready for immediate use, which is ideal for live streaming scenarios.

ComfyUI-FlashVSR Features

  • High-Quality Video Upscaling: The extension can upscale videos to 2x or 4x their original resolution, significantly enhancing clarity and detail.
  • Multiple Model Versions:
  • Full Model: Offers the highest quality output but requires more memory.
  • Tiny Model: Balances speed and quality for faster processing.
  • Tiny Long Model: Optimized for devices with limited memory, suitable for longer videos.
  • SageAttention Optimization: When installed, this feature boosts processing speed by 20-30%. If not installed, the extension defaults to standard processing.
  • Smart Tiling: This feature allows the extension to process videos in smaller sections, making it possible to handle high-resolution videos on devices with limited memory.
  • Automatic Model Download: The necessary models are automatically downloaded from Hugging Face, ensuring you always have the latest version.
  • Audio Passthrough: Maintains the original audio quality and synchronization during video processing.

ComfyUI-FlashVSR Models

The extension supports several models, each tailored for different needs:

  • Wan2_1-T2V-1.1_3B_FlashVSR_fp32.safetensors: The latest model offering improved clarity and stability.
  • Wan2_1-T2V-1_3B_FlashVSR_fp32.safetensors: The previous version of the main model.
  • Wan2.1_VAE.safetensors: Used for video variational autoencoding.
  • Wan2_1_FlashVSR_LQ_proj_model_bf16.safetensors: Designed for low-quality projection.
  • Wan2_1_FlashVSR_TCDecoder_fp32.safetensors: A decoder for the Tiny model. Each model serves a specific purpose, allowing you to choose based on your quality and performance needs.

What's New with ComfyUI-FlashVSR

November 15, 2025

  • FlashVSR 1.1 Model Update: Introduced a new model, Wan2_1-T2V-1.1_3B_FlashVSR_fp32.safetensors, which significantly enhances video clarity, stability, and detail.
  • Frame Duplication Fix: Addressed an issue with frame duplication, improving the overall video quality.

October 24, 2025

  • Initial Release: Launched with FlashVSR and FlashVSR Advanced nodes, automatic model downloading, and support for .safetensors models, audio passthrough, and low-memory tiling.

Troubleshooting ComfyUI-FlashVSR

Here are some common issues and solutions:

  • Missing Model Files: If you encounter a FileNotFoundError for Wan2.1_VAE.safetensors, it may be due to a failed download. To resolve this, delete the FlashVSR folder in ComfyUI/models/ and restart ComfyUI to trigger a re-download.
  • Out of Memory Errors: If you experience memory issues, especially with high-quality presets or 4x upscaling, try using the FlashVSR Advanced node with enable_tiling to reduce memory usage.

Learn More about ComfyUI-FlashVSR

For further learning and support, consider exploring the following resources:

  • FlashVSR Project Page: Offers detailed insights into the FlashVSR model and its capabilities.
  • Hugging Face FlashVSR Models: Access the latest models and updates.
  • Community Forums: Engage with other AI artists and developers to share experiences and solutions. These resources provide valuable information and community support to help you make the most of ComfyUI-FlashVSR.

ComfyUI-FlashVSR Related Nodes

RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.