Download Amuse v3.2.0

Amuse Roadmap

• .NET 10 + OnnxRuntime Latest: Migration to the .NET 10 framework and integration of the latest OnnxRuntime for optimized hardware acceleration. (v3.2.0)
• Python Inference Runtime: Implementation of a dedicated Python backend to provide seamless interop with the broader machine learning ecosystem.
• LoRA Adapters: Support for dynamic loading and blending of LoRA adapters for fine-tuned style and character control.
• Z-Image & FLUX2 Models: Integration of next-generation transformer architectures, including Z-Image and FLUX2, for high-fidelity image synthesis.
• WAN & LTX-2 Video Models: Expansion into generative video using WAN and LTX-2 architectures for advanced motion and temporal consistency.
• STT + TTS Audio Models: Deployment of Speech-to-Text for voice-driven prompting and Text-to-Speech for high-quality audio generation.

---

Required External Plugins

1. ContentFilter add ContentFilter.onnx & ContentFilter.bin to Plugins\ContentFilter
2. CLIPTokenizer add tokenizer files to Plugins\CLIPTokenizer

Note: Easy way is to just install the latest Amuse version and copy the files from the X:\Program Files\Amuse\Plugins directory

---

Required External Licences

ImageSharp v3 Licence Required https://sixlabors.com/pricing/

1. Add licence file to root project directory

FontAwesome Pro v6 Licence Required https://fontawesome.com/v6/download

1. Download the 6.7.2 for the desktop package
2. Add the font files to the Fonts directory
3. Set build action to Resource for all font files

---

Archive:

https://huggingface.co/TensorStack/Amuse