Draw Things is a free native app for Apple devices that runs image and video models directly on Apple Silicon and even on recent iPhones and iPads. It downloads models from its own library (converted to run efficiently on Apple hardware), supports LoRAs, ControlNet, inpainting, outpainting and video, and works fully offline. The core is open source as the community edition; the App Store version is free.

If you have a Mac, iPhone or iPad and want local generation without Python, Draw Things is the easiest and fastest path. It is also unique in running serious models on a phone. Windows and Linux users need another app.

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

macOS 13 or later on Apple Silicon (M1 or newer; Intel Macs are limited), or iPhone and iPad with an A14 chip or newer and iOS 16 or later. 8 GB of memory runs SD 1.5 and SDXL; 16 GB or more is recommended for Flux and video. Model files range from 2 GB to over 20 GB, so free up storage.

Comment obtenir et installer Draw Things

Open the App Store on your Mac, iPhone or iPad and install Draw Things: AI Generation. It is free.
Open the app and tap the model selector at the top; pick a model from the built-in list (for example SDXL, Flux schnell or Wan) and download it inside the app.
To import your own .safetensors files on a Mac, choose Import Model in the model menu and point it at the file; Draw Things converts it.
Optional: install the community edition from GitHub if you want to build from source, or use the gRPC server to run the engine on a Mac and generate from an iPhone.

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 Draw Things

Type your prompt in the text box at the bottom and tap Generate. The image appears on the canvas.
Open the settings panel (the sliders icon) to set the size, steps, guidance scale, sampler and seed. The app suggests sensible defaults per model.
Add a LoRA from the LoRA section in the settings; download it from the list or import a file, set its weight and add the trigger words from the model page.
Paint over part of the canvas with the eraser or brush and generate again to inpaint; move the frame outside the image to outpaint.
Use ControlNet (pose, depth, canny, scribble) from the Control section to guide the composition from a reference image.
Pick a video model (Wan, Hunyuan) and set the number of frames in the settings to generate short clips; export them from the history panel.

Conseils et problèmes courants

On an iPhone, start with SD 1.5 or Flux schnell models at smaller sizes; large models can be slow or run out of memory.
Turn on the Metal FlashAttention and the 8-bit options in the settings to save memory on 8 and 16 GB machines.
The history panel keeps every generation with its full settings; long press to reuse them.
Draw Things has an active Discord and documentation site; model conversion questions are answered there.

Modèles utilisables dans Draw Things

Draw Things runs Stable Diffusion 1.5, SDXL, Flux, Qwen Image, Hunyuan and Wan models and their LoRAs from this catalog; the in-app library covers the popular ones and you can import the rest on a Mac.

Modèles Flux

Modèles Stable Diffusion XL 1.0

Modèles Stable Diffusion 1.5

Modèles Wan

Modèles Qwen

Modèles Hunyuan

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.