ComfyUI Guides
In-depth Windows guides covering VRAM optimization, FP8 and GGUF quantization, workflow JSON sharing, and multi-GPU setups.
Updated Oct 17, 2026 · 3 min read
- Section
- Guides
- Platform
- Windows 10 & 11 (64-bit)
- Supported GPUs
- NVIDIA RTX 20 / 30 / 40 / 50 (CUDA 12.4+)
- Installer
- Available on /download
How ComfyUI Guides works on Windows
In-depth Windows guides covering VRAM optimization, FP8 and GGUF quantization, workflow JSON sharing, and multi-GPU setups. Step-by-step architectural guides for stable generation on 8 GB, 12 GB, 16 GB, and 24 GB NVIDIA RTX cards.
Every page in this directory focuses on practical local execution on Windows 10 and Windows 11, keeping your checkpoints, LoRAs, and generated outputs entirely on your own SSD.
Start by grabbing the current Windows build from /download, or browse the guides and node blueprints linked below.
Key workflow principles
- Keep large `.safetensors` and `.gguf` weights inside organized `models/` subfolders or map shared directories through extra_model_paths.yaml.
- Match your model precision (FP16, FP8 e4m3fn, or GGUF Q4/Q8) to your card's VRAM tier so sampling stays in GPU memory without paging to disk.
- Drag any generated PNG directly onto the ComfyUI Studio canvas to reload its exact node wiring, seed, and sampler values.
Featured pages in ComfyUI Guides
- ComfyUI VRAM Requirements for SDXL, Flux & Video (6GB–24GB) — Direct Windows guide for "comfyui vram requirements": folder paths, node wiring, and VRAM settings in ComfyUI.
- Complete Flux.1 Dev & Schnell Setup Guide in ComfyUI — Direct Windows guide for "comfyui flux setup guide": folder paths, node wiring, and VRAM settings in ComfyUI.
- Hunyuan Video Text-to-Video & Image-to-Video Guide in ComfyUI — Direct Windows guide for "comfyui hunyuan video guide": folder paths, node wiring, and VRAM settings in ComfyUI.
- Wan 2.1 Video Generation Guide in ComfyUI (1.3B & 14B) — Direct Windows guide for "comfyui wan 2.1 video workflow": folder paths, node wiring, and VRAM settings in ComfyUI.
- Running ComfyUI on 8 GB VRAM GPUs (RTX 3060 / 4060) — Direct Windows guide for "comfyui 8gb vram guide": folder paths, node wiring, and VRAM settings in ComfyUI.
- ComfyUI ControlNet Guide: Depth, OpenPose & Canny Wiring — Direct Windows guide for "comfyui controlnet guide": folder paths, node wiring, and VRAM settings in ComfyUI.
- IPAdapter Plus & FaceID Reference Guide in ComfyUI — Direct Windows guide for "comfyui ipadapter guide": folder paths, node wiring, and VRAM settings in ComfyUI.
- Inpainting & Outpainting Workflows in ComfyUI — Direct Windows guide for "comfyui inpainting guide": folder paths, node wiring, and VRAM settings in ComfyUI.
- Two-Pass Hires Fix & Ultimate SD Upscale Guide in ComfyUI — Direct Windows guide for "comfyui hires fix upscaling guide": folder paths, node wiring, and VRAM settings in ComfyUI.
- Staying Safe with ComfyUI Custom Nodes & Safetensors — Direct Windows guide for "are comfyui custom nodes safe": folder paths, node wiring, and VRAM settings in ComfyUI.
- Complete ComfyUI Models Folder Map (Where Every File Goes) — Direct Windows guide for "comfyui models folder structure": folder paths, node wiring, and VRAM settings in ComfyUI.
- ComfyUI on NVIDIA RTX 5090 & 5080 (CUDA 12.8 & FP8) — Direct Windows guide for "comfyui rtx 5090 5080 setup": folder paths, node wiring, and VRAM settings in ComfyUI.
- ComfyUI Wan 2.2 Gguf 8gb — "comfyui wan 2.2 gguf 8gb" is a real ComfyUI search. Install the desktop or portable Windows build, put weights in the matching models folder, then run the graph locally.
- Wan 2.2 Gguf 8gb Vram Workflow — "wan 2.2 gguf 8gb vram workflow" is a real ComfyUI search. Install the desktop or portable Windows build, put weights in the matching models folder, then run the graph locally.
FAQs
Where do I download ComfyUI for Windows to use guides?
Download the Windows desktop installer or portable package directly from /download on this site.
Does ComfyUI run 100% locally without a cloud subscription?
Yes. Once installed on Windows with your local checkpoints in `models/`, ComfyUI executes every node graph locally on your GPU with zero per-image fees.
Run on your hardware. Free forever.
The full ComfyUI engine: open source, fast, extensible, and yours to run however you want. Download the desktop app and open this workflow on your own machine.
