# Comfy > ComfyUI is the node-based AI creation engine for visual workflows. Comfy guides cover ComfyUI, the free and open-source node-based engine for visual AI. Workflows are graphs of connected nodes (model loaders, prompts, samplers, outputs) that can be saved, shared and re-run. The desktop app runs locally on Windows, macOS and Linux. Canonical download: https://comfy.ink/download Last reviewed: 2026-10-17 ## Docs - [ComfyUI Studio Workspace](https://comfy.ink/studio.md): Node-based diffusion studio on Windows: visual graph editor, queue manager, live preview, and local model browser. - [ComfyUI Nodes Reference](https://comfy.ink/nodes.md): Core loaders, KSampler nodes, ControlNet preprocessors, IPAdapter, and ComfyUI Manager custom node packs on Windows. - [ComfyUI Models & Checkpoints](https://comfy.ink/models.md): Where to place safetensors, GGUF, UNet, LoRA, ControlNet, and VAE weights for SDXL, Flux.1, and Hunyuan Video on Windows. - [ComfyUI Guides](https://comfy.ink/guides.md): In-depth Windows guides covering VRAM optimization, FP8 and GGUF quantization, workflow JSON sharing, and multi-GPU setups. - [ComfyUI How-To Tutorials](https://comfy.ink/howto.md): Fast task walkthroughs for installing ComfyUI Manager, fixing missing red nodes, loading PNG metadata, and clearing CUDA OOM. - [ComfyUI vs Other Diffusion Tools](https://comfy.ink/compare.md): Side-by-side comparisons of ComfyUI against Automatic1111, SD WebUI Forge, SwarmUI, Fooocus, InvokeAI, and cloud GPUs. - [ComfyUI Glossary](https://comfy.ink/words.md): Plain-English definitions of every term on the ComfyUI canvas: KSampler, CFG scale, Euler, denoise, VAE, CLIP, and FP8. - [ComfyUI Frequently Asked Questions](https://comfy.ink/ask.md): Direct answers to common questions about ComfyUI Windows requirements, offline operation, commercial rights, and safe weights. - [ComfyUI Desktop for Windows](https://comfy.ink/desktop.md): Everything about the native Windows desktop build: bundled Python 3.12 environment, auto-updater, tray controls, and folder layout. - [ComfyUI Windows Installation](https://comfy.ink/install.md): First-time installation checklists for Windows 10 and 11, PyTorch CUDA 12.4 wheels, Git setup, and shared checkpoint folders. - [ComfyUI Deep-Dive Topics](https://comfy.ink/topics.md): Focused technical notes on low-VRAM flags, Hunyuan Video and Wan 2.1 pipelines, API automation, and batch rendering. - [ComfyUI Model & Task Routes](https://comfy.ink/routes.md): Ready-to-follow node recipes pairing top diffusion models with specific generation tasks and Windows GPU memory tiers. ## Featured - [How to install ComfyUI on Windows](https://comfy.ink/howto/install-comfyui-on-windows.md): Install ComfyUI on Windows with the desktop app: download, run the installer and open your first workflow. - [ComfyUI Studio Workspace](https://comfy.ink/studio.md): Node-based diffusion studio on Windows: visual graph editor, queue manager, live preview, and local model browser. - [ComfyUI Nodes Reference](https://comfy.ink/nodes.md): Core loaders, KSampler nodes, ControlNet preprocessors, IPAdapter, and ComfyUI Manager custom node packs on Windows. - [ComfyUI Models & Checkpoints](https://comfy.ink/models.md): Where to place safetensors, GGUF, UNet, LoRA, ControlNet, and VAE weights for SDXL, Flux.1, and Hunyuan Video on Windows. - [ComfyUI Guides](https://comfy.ink/guides.md): In-depth Windows guides covering VRAM optimization, FP8 and GGUF quantization, workflow JSON sharing, and multi-GPU setups. - [ComfyUI How-To Tutorials](https://comfy.ink/howto.md): Fast task walkthroughs for installing ComfyUI Manager, fixing missing red nodes, loading PNG metadata, and clearing CUDA OOM. - [ComfyUI vs Other Diffusion Tools](https://comfy.ink/compare.md): Side-by-side comparisons of ComfyUI against Automatic1111, SD WebUI Forge, SwarmUI, Fooocus, InvokeAI, and cloud GPUs. - [ComfyUI Glossary](https://comfy.ink/words.md): Plain-English definitions of every term on the ComfyUI canvas: KSampler, CFG scale, Euler, denoise, VAE, CLIP, and FP8. - [ComfyUI Frequently Asked Questions](https://comfy.ink/ask.md): Direct answers to common questions about ComfyUI Windows requirements, offline operation, commercial rights, and safe weights. - [ComfyUI Desktop for Windows](https://comfy.ink/desktop.md): Everything about the native Windows desktop build: bundled Python 3.12 environment, auto-updater, tray controls, and folder layout. - [ComfyUI Windows Installation](https://comfy.ink/install.md): First-time installation checklists for Windows 10 and 11, PyTorch CUDA 12.4 wheels, Git setup, and shared checkpoint folders. - [ComfyUI Deep-Dive Topics](https://comfy.ink/topics.md): Focused technical notes on low-VRAM flags, Hunyuan Video and Wan 2.1 pipelines, API automation, and batch rendering. - [ComfyUI Model & Task Routes](https://comfy.ink/routes.md): Ready-to-follow node recipes pairing top diffusion models with specific generation tasks and Windows GPU memory tiers. - [Flux.1 Dev in ComfyUI](https://comfy.ink/models/flux-1-dev.md): 12B rectified-flow transformer delivering top prompt adherence and typography when loaded via DualCLIPLoader (clip_l + t5xxl) and ae.safetensors. Requires 12 GB+ (8 GB with GGUF Q4 / FP8) VRAM and loads from models/unet/ or models/checkpoints/. - [Stable Diffusion XL 1.0 in ComfyUI](https://comfy.ink/models/sdxl-1-0.md): Native 1024×1024 dual-encoder architecture with the widest ecosystem of LoRAs, ControlNets, and IPAdapter weights. Requires 8 GB+ VRAM and loads from models/checkpoints/. - [Hunyuan Video in ComfyUI](https://comfy.ink/models/hunyuan-video.md): 13B open-weights text-to-video and image-to-video foundation model producing coherent 720p motion sequences inside native ComfyUI video nodes. Requires 12 GB–24 GB (with GGUF / FP8 offload) VRAM and loads from models/diffusion_models/. - [KSampler & KSampler (Advanced)](https://comfy.ink/nodes/ksampler.md): The heart of every ComfyUI graph: iteratively denoises a latent tensor using your chosen solver, scheduler curve, and classifier-free guidance scale. Package: Built-in Core. - [Apply ControlNet (Advanced)](https://comfy.ink/nodes/controlnet-apply-advanced.md): Injects spatial structure from depth maps, OpenPose skeletons, or Canny edges across a chosen percentage window of sampling steps. Package: Built-in Core. - [ComfyUI-Manager](https://comfy.ink/nodes/comfyui-manager.md): Essential management extension that detects missing red nodes in imported workflows and installs custom node repositories with one click. Package: ltdrdata/ComfyUI-Manager. - [ComfyUI vs Automatic1111 SD WebUI](https://comfy.ink/compare/comfyui-vs-automatic1111.md): ComfyUI uses 30–50% less VRAM, launches in seconds, and supports Flux, SD3.5, and Hunyuan Video natively on day one. - [ComfyUI vs SD WebUI Forge](https://comfy.ink/compare/comfyui-vs-sd-webui-forge.md): Forge modernizes the Gradio tab interface with ComfyUI's backend memory management, while ComfyUI gives full modular control over every latent step. ## Optional - [Full catalog](https://comfy.ink/llms-full.txt): every topic page in Markdown - [Sitemap](https://comfy.ink/sitemap.xml)