- Home
- AI Video Upscalers
- FlashVSR
On the recordRead 9 Oct 202613 facts
FlashVSR
Plans and prices
FlashVSR
Free
Open-source code and model weights · Requires a local Python environment and Block-Sparse Attention backend · GPU compatibility varies
Read 9 Oct 2026 · github.com
On the record
- Community integrations
- The README lists community ComfyUI implementations and WanVideoWrapper integration, and says WanVideoWrapper has known issues.github.com · read 9 Oct 2026
- Core methods
- Its framework combines a three-stage distillation pipeline, locality-constrained sparse attention, and a tiny conditional decoder.github.com · read 9 Oct 2026
- Hardware requirements
- The required Block-Sparse Attention backend is reported to work correctly on NVIDIA A100 and A800, while H200 runs with limited acceleration and compatibility on other listed GPUs is unknown.github.com · read 9 Oct 2026
- Install requirements
- The setup instructions use Python 3.11.13 and require installing the Block-Sparse Attention backend.github.com · read 9 Oct 2026
- License
- The GitHub repository identifies its license as Apache-2.0.github.com · read 9 Oct 2026
- Model versions
- The repository provides v1 and v1.1 model weights through Hugging Face, and recommends v1.1.github.com · read 9 Oct 2026
- Performance
- The project reports about 17 FPS for 768 × 1408 video on a single A100 GPU.github.com · read 9 Oct 2026
- Purpose
- FlashVSR is a one-step diffusion framework for streaming video super-resolution.github.com · read 9 Oct 2026
- Quality limitation
- The project says implementations without its sparse attention module may have noticeable quality degradation, especially at higher resolutions.github.com · read 9 Oct 2026
- Recommended setting
- The project is primarily designed and optimized for 4× video super-resolution, which it recommends for better results and stability.github.com · read 9 Oct 2026
- Support
- The maintainers welcome feedback and issues, and list Junhao Zhuang as a contact.github.com · read 9 Oct 2026
- Third-party inference
- The README lists third-party cloud and API deployments and cautions that their model versions or pipeline completeness may differ from the official implementation.github.com · read 9 Oct 2026
- Training dataset
- VSR-120K is described as containing 120,000 videos and 180,000 images for joint image and video training.github.com · read 9 Oct 2026
