This repository gathers custom workflows for ComfyUI, focused on creating generative AI content locally.
These pipelines combine state-of-the-art models and modern techniques (such as Self Forcing, VACE, and LoRAs) — ideal for those working with content automation, research, or creative projects using AI.
Wan 2.1 Self Forcing Video + LoRA + Depth + Image Reference
Fast video generation using a lightweight model (Wan2.1 1.3B Self Forcing), with options for LoRA, depth map, and image reference.
- ComfyUI
- Compatible base models (e.g., VACE, Depth, LoRA, Background Remover)
- Python 3.10+
- GPU with at least 6GB of VRAM
Suggestions, contributions, or questions? Feel free to open an issue or reach out.