This is an early release preview. You may encounter bugs.
SwarmUI logo
Unclaimed

Tool coding

SwarmUI

Modular web UI for AI image and video generation, with an accessible generate tab and a raw Comfy node graph

B 76/100 GitHub score ? This grade is derived from GitHub signals, not user votes. Open for the full breakdown.
No votes yet

01 / About

What SwarmUI is.

SwarmUI, formerly StableSwarmUI, is a modular web interface for AI image, video, and audio generation, written in C# with a JavaScript front end. It runs image models such as Krea 2, Stable Diffusion, and Flux, video models such as MiniMax H3, Wan, and LTX-2, and audio models such as ACE-Step, and adds new model families as they are released publicly.

The interface has two main modes. The Generate tab presents parameters and tooling in a form-based layout with conveniences such as an image editor, automatic workflow generation, and a Grid Generator for batch parameter sweeps. The Comfy Workflow tab exposes the raw ComfyUI node graph for unrestricted control. ComfyUI is auto-installed as the default backend; AUTOMATIC1111's stable-diffusion-webui can be configured as an alternative backend.

The name refers to the project's original function: distributing a single user's generation jobs across a swarm of GPUs, which is used for large grid generations. The project is in beta (v0.9.8); planned work includes mobile browser support, native LLM-assisted prompting (currently an extension), and a desktop distribution. It runs on Windows, Linux, Docker, and Apple silicon Macs, and templates exist for Google Colab, Runpod, and Vast.ai.

Features

  • Model coverage: image (Krea 2, Stable Diffusion, Flux), video (MiniMax H3, Wan, LTX-2), and audio (ACE-Step) models
  • Generate tab: form-based generation with an image editor, auto-generated workflows, and the Grid Generator
  • Comfy Workflow tab: direct editing of the ComfyUI node graph for advanced pipelines
  • Multi-GPU swarm: distributes one user's jobs across several GPUs, including for large grids
  • Backends: ComfyUI installed automatically, with AUTOMATIC1111 stable-diffusion-webui as an option
  • Optional add-ons: YOLOv8 face detection, IP Adapter Face via insightface, aesthetic and PickScore predictors, and rembg background removal
  • Extensions: user-built extensions and installable custom node packs
  • Deployment targets: Windows, Linux, macOS on Apple silicon, Docker, Google Colab, Runpod, and Vast.ai

02 / Discussion CREDIBILITY-GATED

Discussion

Reading is open to everyone. Posting and voting need a verified identity or a GitHub grade of B or higher.

  • No discussions yet.

03 / Build

Build with SwarmUI.

Browse the catalogue for frameworks, tools, and harnesses, each scored on real GitHub credibility.

Get SwarmUI →

Browse the catalogue