SwarmUI (formerly StableSwarmUI) is a free, open source web interface that uses ComfyUI as its backend. You get a clean Generate tab with a prompt box, parameter sliders and a model browser with previews, while the Comfy Workflow tab lets you open the underlying graph for anything the simple tab cannot do. It supports images and video, multiple GPUs and even several machines at once, hence the name.

SwarmUI is ideal if you want ComfyUI's model support (Flux, Wan, Hunyuan, Qwen, Chroma and more) with a normal interface. It is also a good choice for people who switch between quick generation and detailed workflows. The installer takes care of ComfyUI for you.

Aucune installation requise – exécutez les modèles IA dans le cloud

BitVector Prism est la façon la plus simple de commencer : choisissez un modèle, tapez un prompt et générez dans une application web épurée, sans rien configurer. BitVector est aussi disponible sur Discord et sur le web (SpyGlass).

Essayer BitVector

Ce qu’il vous faut

Windows 10 or 11, Linux or macOS (Apple Silicon). NVIDIA GPU recommended (8 GB VRAM for SDXL, 12 GB or more for Flux and video); AMD on Linux with ROCm, or on Windows with some setup. .NET 8 runtime is installed by the installer. Disk space as for ComfyUI: tens of gigabytes for models.

Comment obtenir et installer SwarmUI

Download the Windows installer (install-windows.bat) from the GitHub README or the site, put it in the folder where you want SwarmUI, and run it. Linux and Mac have equivalent install scripts.
The installer opens a browser page: choose Just Install Everything for a fresh ComfyUI backend, or point it at an existing ComfyUI if you have one.
Pick a starter model to download during setup (SDXL, Flux or SD 1.5), or skip and add your own later.
Put your own models in Models/Stable-Diffusion (checkpoints and Flux), Models/Lora, Models/VAE and Models/clip. SwarmUI can also read the folders of an existing ComfyUI install.
Start SwarmUI with launch-windows.bat (or the Linux/Mac launch script). The interface runs at http://localhost:7801.

Téléchargez uniquement depuis les liens officiels de cette page. Les installateurs trouvés ailleurs peuvent être obsolètes ou modifiés. ModelTrends.ai n’héberge aucun fichier.

Aucune installation requise – exécutez les modèles IA dans le cloud

PirateDiffusion est réservé à Telegram et conçu pour les pros : des milliers de modèles, LoRA et workflows pilotés par commandes de chat, avec génération illimitée à prix fixe.

Essayer PirateDiffusion

Comment utiliser SwarmUI

Click the Models tab at the bottom, pick a model (hover for its preview and description) and it becomes the active model.
Write your prompt in the box, set Images, Steps, CFG Scale, Resolution and Seed in the left panel, and press Generate.
Add a LoRA by selecting it in the LoRAs tab at the bottom, then set its weight in the prompt tag it creates. Remember the trigger words from the model page.
Use the Init Image section for image to image, the Refiner section for a second pass, and ControlNet for guided generation; everything is a checkbox in the left panel.
Open the Comfy Workflow tab to see and edit the node graph behind your current settings, or to load any ComfyUI workflow.

Conseils et problèmes courants

Edit a model's metadata (right click in the Models tab) to save its trigger words and default settings; they are applied automatically.
Use the Grid Generator tool to compare several models, LoRAs or settings in one run.
If the backend fails to start, open the Server > Backends tab: it shows the ComfyUI log and lets you restart it.
Video models (Wan, Hunyuan, LTX) appear in the Models tab like any other model; set the frame count in the Video section.

Modèles utilisables dans SwarmUI

Through its ComfyUI backend, SwarmUI runs almost everything in this catalog: Stable Diffusion 1.5, SDXL, Flux, Flux 2, Qwen Image, Chroma, Hunyuan, Wan and their LoRAs.

Modèles Flux

Modèles Flux 2 / Klein

Modèles Stable Diffusion XL 1.0

Modèles Stable Diffusion 1.5

Aucune installation requise – exécutez les modèles IA dans le cloud

Pas de GPU, pas de téléchargement, pas de Python. Ces services exécutent les mêmes modèles open source sur leurs propres serveurs : vous générez depuis un navigateur, Discord ou Telegram en une minute.

Essayer BitVector

Essayer PirateDiffusion

Liens

Ces liens mènent aux sites des éditeurs. ModelTrends.ai n’est pas affilié à eux et n’héberge aucun logiciel.

Autres applis locales

Comment installer ComfyUI

The node-based powerhouse for Stable Diffusion, Flux, Wan video and almost every new open source model. Free, open source, for Windows, Mac and Linux.