Amuse is a free Windows application made with AMD for local image generation. It is a normal installed program with a gallery, a text to image screen, image to image, inpainting, upscaling and a video section, and it downloads ready-made models from its own library. It uses ONNX and DirectML, so it runs on AMD Radeon graphics cards and Ryzen AI laptops without special drivers, and also on other Windows GPUs.

Amuse is the easiest option for Windows users with an AMD card, who otherwise struggle with the NVIDIA-first tools. Advanced users will miss custom checkpoints in arbitrary formats and the extension ecosystem of other apps.

No install required - run AI models on the cloud

BitVector Prism is the easiest way to start: pick a model, type a prompt and generate in a clean web app, with nothing to configure. BitVector is also available on Discord and on the web (SpyGlass).

Try BitVector

What you need

Windows 10 or 11 (64-bit). AMD Radeon RX 7000 or 6000 series, Ryzen AI (XDNA) laptops or other DirectML-capable GPUs with 8 GB or more VRAM recommended for SDXL; 16 GB system RAM. Models are downloaded in the app, several gigabytes each.

How to get and install Amuse

Download the Amuse installer from amuse-ai.com and run it.
Open Amuse, accept the licence and choose the mode (EZ mode for a guided experience, Expert for all settings).
Open the Models screen and download a model from the list (Stable Diffusion 1.5, SDXL, SDXL Turbo, Flux and others, already converted for Amuse).
Optional: in Expert mode, import your own ONNX converted models through the Add Model option.

Download only from the official links on this page. Installers found elsewhere may be out of date or tampered with. ModelTrends.ai hosts no files.

No install required - run AI models on the cloud

PirateDiffusion is Telegram only and built for pros: thousands of models, LoRAs and workflows driven by chat commands, with unlimited generation on a fixed price plan.

Try PirateDiffusion

How to use Amuse

Choose Text to Image, pick a model, type your prompt and press Generate.
In Expert mode set the steps, guidance, size, scheduler and seed, and add a negative prompt.
Use Image to Image to transform a loaded picture, Inpaint to paint a mask and regenerate it, and the Upscale tool to enlarge results.
Pick a LoRA in the model options if the app has one for your base model, and add the trigger words from the model page.
Finished pictures are in the Gallery with their settings; right click to save or reuse.

Tips and common problems

Keep your Radeon driver current; DirectML performance and model support improve with driver updates.
Turbo and Lightning models produce images in a few steps and are the fastest option on mid-range cards.
If a download stalls, pause and resume it in the Models screen rather than restarting the app.
Amuse cannot load .safetensors files directly; use the models from its library or convert to ONNX first.

Models you can run in Amuse

Amuse runs the Stable Diffusion 1.5, SDXL and Flux models offered in its library, which correspond to base models in this catalog. Custom checkpoints need ONNX conversion.

Stable Diffusion XL 1.0 Models

Stable Diffusion 1.5 Models

Flux Models

No install required - run AI models on the cloud

No GPU, no downloads, no Python. These services run the same open source models on their own servers, so you can start generating from a browser, Discord or Telegram in a minute.

Try BitVector

Try PirateDiffusion

Links

These links go to the makers' own sites. ModelTrends.ai is not affiliated with them and does not host any software.

Other local apps

How to install 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.