ComfyUI
Node-based visual AI image pipeline. Compose complex generation workflows visually.
One-time setup
What is this?
Visual AI image generation, reinvented.
ComfyUI is a node-based visual interface for building AI image generation workflows. Unlike traditional UIs with simple text boxes, ComfyUI lets you visually connect processing nodes to create complex pipelines — giving you unprecedented control over every step of the generation process.
Each node represents a distinct step in the pipeline: text encoding, model loading, sampling, conditioning, and decoding. You drag, drop, and connect them to build exactly the workflow you need — from simple text-to-image to multi-stage inpainting with ControlNet guidance.
ComfyUI runs entirely locally on your Mac's Apple Silicon, leveraging MPS acceleration for fast generation without sending any data to the cloud. Your prompts, images, and creative process stay completely private.
Whether you are a professional artist building production pipelines or an enthusiast exploring AI art, ComfyUI's modular approach lets you iterate faster and achieve results that traditional interfaces simply cannot match.
How it works
From node to masterpiece.
Text / Image] --> B[CLIP Encoder] B --> C[KSampler] C --> D[Model Node
SDXL / Flux] D --> E[VAE Decode] E --> F[Preview /
Save Node]
What you get
Everything included in this service.
- ComfyUI installed and configured
- SDXL and Flux model support
- Custom workflow templates
- Node-based visual pipeline builder
- ControlNet integration
- LoRA support for fine-tuned styles
- Optimised for Apple Silicon
- Quick-start workflow library
Who is this for?