GitHub - comfyanonymous/ComfyUI: The most powerful and modular stable diffusion GUI, api and backend with a graph/nodes interface.
GitHub - comfyanonymous/ComfyUI: The most powerful and modular stable diffusion GUI, api and backend with a graph/nodes interface.https://github.com/comfyanonymous/ComfyUI

GitHub - comfyanonymous/ComfyUI: The most powerful and modular stable diffusion GUI, api and backend with a graph/nodes interface.

🚀 Explore ComfyUI: A powerful GUI & backend tool for intricate workflows without coding! 💻 Create stable diffusion workflows effortlessly using graph/nodes interface. 🌟 Supports GPU/CPU ops & advanced models like VAEs & CLIP. 🎨 Perfect for Hires fix & image processing! #AI #GUI

  • ComfyUI is a robust GUI and backend for stable diffusion workflows.
  • It offers a graph/node interface for creating intricate workflows without coding.
  • Supports SD1.x, SD2.x, SDXL, Stable Video Diffusion, and Stable Cascade workflows.
  • Features asynchronous queue system and optimization for re-executing changed parts only.
  • Supports GPU and CPU operation, with low VRAM mode auto-enabled on GPUs with < 3GB VRAM.
  • Allows loading ckpt, safetensors, and diffusers models/checkpoints.
  • Provides various models like VAEs, CLIP models, embeddings, and textual inversion.
  • Offers features like Area Composition, Inpainting, Loras, ControlNet, and T2I-Adapter.
  • Facilitates the creation of high-quality workflows for tasks like Hires fix and image processing.
  • Enables standalone VAEs and CLIP models, with support for multiple advanced models.
  • Operates fully offline without any downloads required.
  • Provides configurable search paths for models and JSON-based workflow saving/loading.
  • Offers various shortcuts for workflow manipulation and management.
  • Accommodates various platforms like Windows, Linux, Intel Arc, AMD GPUs, and Apple Mac silicon.