ComfyUI

ComfyUI is an open-source, node-based interface for generating images and video locally with Stable Diffusion and other models, each step of the workflow a node on a canvas.

Comfy is an AI creation engine designed for visual professionals who demand control over every model, parameter, and output. It enables users to connect models, processing steps, and outputs on an infinite canvas where every decision is visible and inspectable. Users can start from community templates or build workflows from scratch, with an App mode offering a simplified view. The platform supports thousands of community-shared workflows and nodes for tasks like image and video generation. It runs locally via Comfy Desktop or in the cloud through Comfy Cloud, and offers a developer platform for production APIs.

ComfyUI at a glance

Best for: Designer, Social media creator, Software engineer
Runs on: Web, Desktop, Cloud, API, self-hosted
Pricing: Free plan with 5 GPU runs; Standard, Creator, Pro, and Team paid plans available with monthly credits and hosted models.
Released: January 2023
Category: AI image tool
Website: comfy.org

Key features

Connect models and steps on an infinite canvas
Switch between App mode and node graph view
Browse and remix community workflows
Deploy workflows as a production API
Run locally via Comfy Desktop
Access cloud-powered workflows with Comfy Cloud
Import custom models and LoRAs

Use it for

Generate AI images and videos using nodes
Build and share custom visual AI workflows
Deploy production APIs for workflows
Train and run models locally on hardware

Companions

Tools that articles about ComfyUI recommend using with it, each with its job.

ComfyUI Manager — Install and update every other custom node from inside ComfyUI.
ComfyUI Impact Pack — Face detection, segmentation and detail fixing.
ControlNet Aux — Preprocessors for pose, depth and edge control.
Ultimate SD Upscale — High-resolution tiled upscaling.
VideoHelperSuite — Loads and combines video frames for animation workflows.
rgthree-comfy — Quality-of-life nodes that tidy up big workflows.

Companion to

Tools whose articles name ComfyUI as a companion, each with its job for them.

Flux — ComfyUI: Workflow
Higgsfield — ComfyUI: Backup pipeline
Open WebUI — ComfyUI: Image generation

Alternatives to ComfyUI

Modly — Generates 3D models locally on your GPU for free without cloud processing.
LayerShop — Generative AI built into a native image editor with local model execution.
SwarmUI — It offers modular powertools and high performance for managing multiple AI models.
Civitai — Biggest library of community-made models and LoRAs
Leonardo — Generates images and videos with custom AI models and real-time canvas tools.
Krea — Every major image and video model plus a 21K enhancer in one tool
Flux — Runs via API or as open weights on your own infrastructure.
Lovart — Generates a full marketing campaign kit in one canvas from a single prompt.
Pollinations — Build AI applications quickly with a single API and earn rewards.
Invoke — Offers an industry-leading web UI and a fully integrated unified canvas.
getimg.ai — Generates images, videos, music, and speech using leading AI models in one subscription.
Photo Ferret — Browse large photo collections with sliders and tags instead of scrolling.

All ComfyUI alternatives