SwarmUI
SwarmUI, formerly StableSwarmUI, is an MIT-licensed web interface for AI image and video generation that aims to make power tools easy to reach. A simple Generate tab suits beginners, while a Comfy Workflow tab gives the full ComfyUI graph, which it can install as its backend. It supports Stable Diffusion, Flux, Wan and LTX, and can spread generation across a swarm of GPUs. It is free with no paid tier.
What does SwarmUI do?
SwarmUI is "a modular AI image generation web-user-interface, with an emphasis on making powertools easily accessible, high performance, and extensibility." It supports image models such as Krea 2, Stable Diffusion and Flux, video models such as MiniMax H3, Wan and LTX-2, and some audio models such as ACE-Step. The README recommends it "as an ideal UI for most users, beginners and pros alike."
It has two interfaces over one engine: a form-based Generate tab, and a Comfy Workflow tab "to get the unrestricted raw graph". It can install ComfyUI automatically as its backend and can also use AUTOMATIC1111. The name refers to letting a swarm of GPUs generate images for one user at once. It includes an image editor, automatic workflow generation, a grid generator for comparing settings, face IP-Adapter support and YOLO face detection. The README says it is 100 percent free and open source, funded by donations.
Key facts
- Licence: MIT, copyright Alex Goodwin; updates before June 2024 were MIT, copyright Stability AI. The README notes some uses may fall under GPL-family licences of connected projects, and optional YOLO face detection may bring AGPL terms.
- The README states "any models used have their own licenses."
- Models: images such as Krea 2, Stable Diffusion and Flux; video such as MiniMax H3, Wan and LTX-2; some audio such as ACE-Step.
- A form-based Generate tab and a raw Comfy Workflow tab; ComfyUI installs automatically as the backend, with AUTOMATIC1111 as an option.
- Multi-GPU generation for one user, the swarm in its name.
- Image editor, auto-workflow generation, grid generator, face IP-Adapter and YOLOv8 face detection.
- Install: installer scripts for Windows and Linux, a launch script for macOS on M-series, Docker, and Colab; needs git, .NET 8 SDK and Python 3.10 to 3.12.
- Status: "Almost-Release" per the README. Latest tagged release when read: 0.9.8-Beta on 2026-02-06, with pushes through 2026-09-17; 4,571 stars.
What does it replace, and where does it fall short?
SwarmUI replaces Midjourney for someone who wants a prompt box that just works today and a path to full control tomorrow, since the same install gives both a simple form and ComfyUI's graph. Its multi-GPU support suits a studio with several cards. It is the pick for beginners who want room to grow on the open-source alternatives to Midjourney page.
Where it falls short: the README calls it "Almost-Release" and its latest tag is a beta from February 2026. LLM-assisted prompting is not implemented yet. The Windows installer sometimes needs running twice, and the desktop app mode is untested on Linux. For the engine underneath, ComfyUI.
How does SwarmUI run?
On Windows run install-windows.bat, on Linux install-linux.sh, on an M-series Mac launch-macos.sh, or use Docker. It needs git, the .NET 8 SDK and Python 3.10 to 3.12, and serves on port 7801. A Colab notebook and third-party RunPod and Vast.ai templates exist for cloud GPUs. There is no paid tier; it is funded by Patreon donations.
Who is SwarmUI for?
A beginner who wants simple image generation now and ComfyUI's full power later, or a studio spreading work across several GPUs. Someone who wants a canvas for editing should use InvokeAI.
What limits does the README state?
From the README: "This project is in Almost-Release status"; LLM-assisted prompting is not yet implemented; the Windows installer "sometimes needs to run twice"; desktop app mode is not tested on Linux; Colab may not allow remote web interfaces on free accounts.
Questions people ask
Is SwarmUI open source?
Yes. SwarmUI's own code is MIT, and the README says it is 100 percent free and open source forever. It notes that some uses may fall under GPL-family licences of connected projects such as ComfyUI, that optional YOLO face detection may bring AGPL terms, and that every model has its own licence.
What is the difference between SwarmUI and ComfyUI?
SwarmUI is a web interface that can install and drive ComfyUI as its backend. It adds a simple Generate tab, an image editor, a grid generator and multi-GPU support, while its Comfy Workflow tab gives the raw ComfyUI graph when you need it.
Can SwarmUI use several GPUs?
Yes. The name comes from letting a swarm of GPUs generate images for the same user at once. That is its main difference from running ComfyUI or InvokeAI on a single card.
How does SwarmUI compare with Midjourney?
SwarmUI gives a simple prompt interface over open models on your own hardware, with ComfyUI's full graph available behind it, for free. Midjourney is a hosted service. The open-source alternatives to Midjourney page compares it with ComfyUI, InvokeAI and SD.Next.
Is SwarmUI finished?
Not by its README, which describes it as in Almost-Release status and lists LLM-assisted prompting as a key feature not yet implemented. Its latest tagged release is 0.9.8-Beta from February 2026, though the repository receives regular pushes.
Sources
- SwarmUI README and MIT LICENSE: github.com/mcmonkeyprojects/SwarmUI, read 2026-09-17.
Compared with the others
On the open-source alternatives to Midjourney page, SwarmUI is the pick for a simple tab now, a graph later. MIT interface with a beginner Generate tab and the raw ComfyUI graph behind it, plus multi-GPU generation for one user.
Also on that page: ComfyUI for control, and the newest models, InvokeAI for a canvas for inpainting and editing, SD.Next for modest or unusual hardware.
More image and video generation
The GPL-3.0 node-graph engine for image, video, audio and 3D generation that supports the newest open models, from Stable Diffusion and Flux to Wan and LTX video, on NVIDIA, AMD, Intel and Apple hardware, offline.
An Apache-2.0 creative image tool with a unified canvas for inpainting and outpainting, node workflows, and support for 28 model families from Stable Diffusion to Flux and Qwen Image, run as a local web app.
An Apache-2.0 all-in-one web interface for AI image and video generation, descended from AUTOMATIC1111's WebUI, with LoRA, ControlNet, quantisation that cuts video memory up to four times, and the widest hardware list here.
Added September 17, 2026. Every claim above comes from the project's README, LICENSE or model card, read on September 17, 2026, or from the GitHub API on the date shown in the panel. Found an error? Write to reveneau@licheo.com and it is fixed in the next weekly pass. Repository: github.com/mcmonkeyprojects/SwarmUI.
Running one of these inside your own environment, with your own data and your own security rules, is the kind of work Reveneau does. Read how a forward deployed engagement works.