Train SD 1.5 and SDXL LoRAs in ComfyUI with the LarryJane491 custom node.
ComfyUITemplates.com
Discover free ready-made ComfyUI templates for AI workflows.
IPAdapter Plus style transfer. One reference image in, prompt drives content. Official example.
Depth ControlNet via FLUX.1 Depth LoRA. Foreground/midground/background stay locked.
Canny edge ControlNet with FLUX.1. Auto-extracts edges, locks composition, prompt drives style.
2-pass OpenPose ControlNet. AnythingV3 base, AOM3A3 refiner, pose-locked from a reference image.
Qwen 2511 multi-reference image editor. 12GB GGUF runs on RTX 3060, 1328x1328 native.
Two-pass detailer for rendered images. Recovers faces, hands, and small text without a reroll.
Fast interactive video character replacement: upload, mark two points, pick a reference.
Artifact-free native 4K+ image generation with FLUX via direct UNet patching through DyPE.
Wan 2.2 text-to-image with fast render and cinematic color science for vivid film-like output.
WAN 2.2 image-to-video with ultra-smooth motion, smart prompt extension, cinematic color.
Swap any character in a video with a reference image and two points; Reactor refines the face.
Filters
Tags
Free ComfyUI Workflow Templates for Image, Video, Inpainting & Upscaling
ComfyUI Templates is a free, no-signup library of pre-built ComfyUI workflow templates you can drop straight into ComfyUI. Every template is a real node graph — load the JSON or PNG into ComfyUI and the full pipeline appears, ready to run or tweak.
What you'll find here
We curate workflow templates across the categories ComfyUI users actually reach for:
- Text-to-Image — SDXL, Flux, and Wan prompt-to-image pipelines with checkpoint + sampler + negative prompt controls.
- Image-to-Image — VAE-encoded restyles, pose transfers, and structural edits on a reference image.
- Inpainting — mask-driven area edits for object removal, detail fixes, and targeted regenerations.
- Image Upscaling — ESRGAN, pixel resampling, and SD secondary-sampling passes.
- Video Generation & Upscaling — node-based diffusion pipelines for animation, interpolation, and frame-by-frame upscale.
- Style Transfer — model-driven artistic re-processing of existing images.
- Background Editing — subject isolation, background replacement, and contextual background generation.
Why templates instead of a blank canvas
ComfyUI is powerful, but wiring a multi-node graph from scratch eats hours the first time. Templates give you a working pipeline on day one — you swap models, prompts, or inputs and iterate, instead of debugging whether your KSampler settings are the reason the output looks wrong.
Every template is free, released under permissive licenses, and hosted on the ComfyUITemplates.com listings below. Drag the JSON or PNG into ComfyUI, queue the prompt, and watch the graph run.
Getting started
- Pick a template from the listings on this page.
- Download the JSON or PNG (the PNG embeds the workflow — drop it into ComfyUI to load the graph).
- Open ComfyUI, load the template, and queue the prompt with the default inputs.
- Swap in your own checkpoint, prompt, or input image to customise.
Browse the listings below to find a workflow for what you want to make. New templates are added as new ComfyUI capabilities ship.