DiffusionBee was the first drag-and-drop app that let Mac users run Stable Diffusion without any setup. It is a normal desktop application with a text to image screen, image to image, inpainting, outpainting, upscaling and a built-in model downloader. Everything runs locally and offline. Newer versions added Flux and SDXL and a Windows build.

DiffusionBee is for people who want the simplest possible start on a Mac and do not need advanced controls. If you later want LoRAs with fine control, ControlNet or video, Draw Things or ComfyUI go further.

Sem instalação: rode modelos de IA na nuvem

BitVector Prism é a forma mais fácil de começar: escolha um modelo, digite um prompt e gere em um app web limpo, sem configurar nada. O BitVector também está disponível no Discord e na web (SpyGlass).

Experimentar o BitVector

O que você precisa

macOS 13 or later; Apple Silicon (M1 or newer) with 16 GB RAM recommended for SDXL and Flux, 8 GB works for SD 1.5. Intel Macs run very slowly. The Windows build needs a 64-bit system and an NVIDIA GPU. Models are 2 to 12 GB each.

Como obter e instalar o DiffusionBee

Go to diffusionbee.com and download the version for your Mac (Apple Silicon or Intel) or for Windows.
Open the .dmg and drag DiffusionBee into Applications, then start it. The first start downloads the default model.
Open the Models screen to download more models (SDXL, Flux) from inside the app, or use Add New Model to import a .safetensors or .ckpt file from Civitai or Hugging Face.
Optional: import LoRAs in the same screen under the LoRA section; they become selectable in the generate options.

Baixe apenas pelos links oficiais desta página. Instaladores de outros lugares podem estar desatualizados ou adulterados. O ModelTrends.ai não hospeda arquivos.

Sem instalação: rode modelos de IA na nuvem

O PirateDiffusion é exclusivo do Telegram e feito para profissionais: milhares de modelos, LoRAs e workflows por comandos de chat, com geração ilimitada a preço fixo.

Experimentar o PirateDiffusion

Como usar o DiffusionBee

On the Generate screen, type your prompt, choose the model from the dropdown and press Generate.
Open Options to set the number of images, steps, guidance scale, size and seed, and to add a negative prompt.
Use Image to Image to transform a picture you drop in, with a strength slider; Inpainting and Outpainting let you paint a mask or extend the canvas.
Pick a LoRA in the options and add its trigger words from the model page to your prompt.
Finished images are in the History screen; click one to save it or reuse its settings.

Dicas e problemas comuns

Close other heavy apps while generating on 8 GB Macs; memory pressure slows generation a lot.
If an imported model does not load, make sure it is a standard SD 1.5, SDXL or Flux checkpoint and not a LoRA or a merged file in an unusual format.
Upscale results with the built-in upscaler instead of generating at large sizes; it is faster and uses less memory.
Check the site for updates; model support (Flux, SDXL) arrived through app updates.

Modelos que você pode rodar no DiffusionBee

DiffusionBee runs Stable Diffusion 1.5, SDXL and Flux checkpoints and LoRAs from this catalog through its import feature.

Modelos Flux

Modelos Stable Diffusion XL 1.0

Modelos Stable Diffusion 1.5

Sem instalação: rode modelos de IA na nuvem

Sem GPU, sem downloads, sem Python. Esses serviços rodam os mesmos modelos de código aberto em seus próprios servidores, então você começa a gerar pelo navegador, Discord ou Telegram em um minuto.

Experimentar o BitVector

Experimentar o PirateDiffusion

Links

Estes links levam aos sites dos próprios criadores. O ModelTrends.ai não é afiliado a eles e não hospeda software.

Outros apps locais

Como instalar o 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.