adrientoupet/SeedVR2_comfyUI

SeedVR2 Test Models

This repository contains test models for SeedVR2 video upscaling. These are experimental versions being tested before merging back into the main numz/SeedVR2_comfyUI repository.

🚧 Development Notice

⚠️ These are TEST MODELS and may be unstable or have unexpected behavior. For production use, please use the official models from numz/SeedVR2_comfyUI.

📋 About SeedVR2

SeedVR2 is a high-quality video upscaling model that uses diffusion techniques to enhance video resolution while maintaining temporal consistency. This implementation is optimized for ComfyUI and supports both standalone and multi-GPU operation.

🔗 Related Projects

• Official SeedVR2 by ByteDance: ByteDance-Seed/SeedVR
• ComfyUI Implementation: numz/ComfyUI-SeedVR2_VideoUpscaler
• Official HuggingFace Models: numz/SeedVR2_comfyUI
• Tutorial by AInVFX: SeedVR2 Deep Dive Tutorial

🛠️ Testing Guidelines

When testing these models:

1. Compare output quality against official models
2. Monitor VRAM usage and performance
3. Test various input resolutions and batch sizes
4. Report issues on the main repository

📄 License

The code is released under the MIT License. Model weights are released under Apache 2.0 License.

🙏 Credits

• Original SeedVR2 implementation by ByteDance Seed
• ComfyUI implementation and optimization by numz
• BlockSwap integration by Adrien Toupet from AInVFX

📞 Support

For questions and support:

• Use the main repository issues
• Check the comprehensive tutorial by AInVFX

---

Remember: These are experimental test models. Always backup your work and test thoroughly before using in production workflows.