Amuse

Amuse is a Windows app for people who want to generate media and text with models running on their own hardware. It combines generation with editing tools for both generated output and existing local video.

• Image generation and editing Image pipelines include FLUX.1, FLUX.2, Qwen and StableDiffusion-XL, with inpainting, selection and masking tools, plus LoRA and ControlNet support for output control.
• Video creation and editing Wan and CogVideoX support video generation, while the video editor accepts generated or local content.
• Upscaling and frame interpolation You can upscale images and video, increase video frame rates or create slow-motion footage.
• Audio and text models Supertonic generates voices, Whisper recognizes speech, ACE-Step XL and LongCat Audio support audio generation, and Qwen and Gemma support text generation.
• Media organization A gallery lets you organize and manage media within the application.

Amuse accepts Safetensors, GGUF and ONNX models. NVIDIA GPUs can use CUDA acceleration; compatible AMD, NVIDIA and Intel GPUs can use Vulkan. NVIDIA RTX cards are recommended for generation performance, and actual compatibility depends on the GPU, drivers, operating system and backend.

The source code is open source under the Apache 2.0 license.