---
title: FlashVSR
canonical_url: "https://www.modelscope.cn/models/1038lab/FlashVSR"
md_url: "https://www.modelscope.cn/models/1038lab/FlashVSR.md"
repository: 1038lab/FlashVSR
last_updated: 2025-12-30
license: apache-2.0
parameters: 3.6B
tensor_type:
  - F32
  - BF16
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 4194
stars: 5
---

# FlashVSR

> FlashVSR - 1038lab 在 ModelScope 开源的模型。🧠 FlashVSR (ComfyUI Integration Model)

1038lab/FlashVSR 是 ModelScope 魔搭社区上的 3.6B 参数机器学习模型，采用 apache-2.0 许可。

- **Repository**: 1038lab/FlashVSR
- **License**: apache-2.0
- **Parameters**: 3.6B
- **Downloads**: 4194
- **Stars**: 5
- **Last updated**: 2025-12-30

Source: https://www.modelscope.cn/models/1038lab/FlashVSR

---

## 🧠 FlashVSR (ComfyUI Integration Model)

This repository hosts the **FlashVSR model weights** used in the [ComfyUI-FlashVSR](https://github.com/1038lab/ComfyUI-FlashVSR) project.
It provides fast and efficient **video super-resolution** powered by **FlashAttention-based architectures**.

### 📦 Model Overview

**FlashVSR** (Flash Video Super-Resolution) is a high-performance video SR model that leverages efficient attention computation to upscale low-resolution videos while maintaining temporal consistency.

This version is provided for **ComfyUI integration**, optimized for direct use with the [ComfyUI-FlashVSR extension](https://github.com/1038lab/ComfyUI-FlashVSR).

### 🧩 Model Files

| File                                             | Description                  | Size    |
| ------------------------------------------------ | ---------------------------- | ------- |
| `Wan2_1-T2V-1_3B_FlashVSR_fp32.safetensors`      | Main FlashVSR model weights  | ~5.7 GB |
| `Wan2_1_FlashVSR_TCDecoder_fp32.safetensors`     | Temporal consistency decoder | ~181 MB |
| `Wan2_1_FlashVSR_LQ_proj_model_bf16.safetensors` | Low-quality projection model | ~576 MB |
| `Wan2.1_VAE.safetensors`                         | VAE weights                  | ~254 MB |
| `Prompt.safetensors`                             | Conditioning embeddings      | ~4 MB   |

### ⚙️ Usage (with ComfyUI)

1. **Install** [ComfyUI](https://github.com/comfyanonymous/ComfyUI)
2. **Clone** the FlashVSR node extension:

   ```bash
   git clone https://github.com/1038lab/ComfyUI-FlashVSR.git
   ```
3. **Download the model files** from this Hugging Face repo:

   * Place them inside:

     ```
     ComfyUI/models/FlashVSR/
     ```
4. **Launch ComfyUI** and load the FlashVSR workflow.

### 🧑‍🔬 Original Work

This model is based on the research and implementation from
👉 [JunhaoZhuang/FlashVSR](https://huggingface.co/JunhaoZhuang/FlashVSR)
All credit for the model architecture and original training goes to the authors.

This repository only repackages the model for **ComfyUI integration**.

### 📜 License

Please refer to the original [FlashVSR license](https://huggingface.co/JunhaoZhuang/FlashVSR) for usage rights.
Any redistribution or fine-tuning should comply with the same terms.

### 🧰 Related Resources

* 🧩 [ComfyUI-FlashVSR GitHub Repo](https://github.com/1038lab/ComfyUI-FlashVSR)
* 📚 [FlashVSR Original Model](https://huggingface.co/JunhaoZhuang/FlashVSR)
* 💬 [Hugging Face Discussion](https://huggingface.co/1038lab/FlashVSR/discussions)