Releases: TensorStack-AI/Diffuse
v0.4.0 - Memory Modes, Quantization & ROCm 7.2
Memory Modes, Quantization & ROCm 7.2
This release adds memory management and quantization options, giving more control over performance and VRAM usage across different hardware configurations.
Features
-
INT8 Quantization Support
Reduce VRAM usage by up to ~30–40% with automatic INT8 quantization during model load. Quantization is applied transparently and requires no manual configuration. -
Memory Modes
New memory modes simplify how models are distributed across GPU(s) and CPU memory, ranging from maximum performance to minimum VRAM usage. This makes Diffuse more accessible on low-memory systems while still scaling well on high-end hardware. -
ROCm 7.2 Support
Added support for ROCm 7.2, improving compatibility and stability on supported AMD GPUs. -
Bug Fixes & Stability Improvements
Various fixes and internal cleanups to improve reliability during model loading and inference.
Full Changelog: v0.3.8...v0.4.0
Installation
1. Installer Version
- Uninstall Diffuse v0.3.X
- Download and run Diffuse_v0.4.0.exe
- Follow the on-screen instructions
2. Standalone Version
-
Download and extract Diffuse_v0.4.0.zip
A fast SSD with plenty of free space is recommended, as model downloads can be large. -
Run Diffuse.exe
Diffuse will automatically:
- Install an isolated portable Python runtime
- Create the required virtual environment
- Download the selected model from Hugging Face
First-run notice
On first launch or when loading a model for the first time, setup may take several minutes while Python, dependencies, and model files are downloaded and initialized. This is expected behavior.
No manual Python setup is required.
Device Support
Supports CUDA and ROCM based devices
Important
Important: This release includes many large changes. A full uninstall and reinstall is recommended (your downloaded models are safe to keep).
This version will overwrite your existing Settings.json. If you have added custom environments or models, you may want to back up this file before upgrading.
v0.3.8 - Add Model Wizard
Model Wizard
Add local Models/LoRA from Civitai or Huggingface easily using the new Model Wizard
Features
- Add Model Wizard
- Improved scheduler support
- Other bug-fixes
Full Changelog: v0.3.7...v0.3.8
Installation
1. Installer Version
- Uninstall Diffuse v0.3.7
- Download and run Diffuse_v0.3.8.exe
- Follow the on-screen instructions
2. Standalone Version
-
Download and extract Diffuse_v0.3.8.zip
A fast SSD with plenty of free space is recommended, as model downloads can be large. -
Run Diffuse.exe
Diffuse will automatically:
- Install an isolated portable Python runtime
- Create the required virtual environment
- Download the selected model from Hugging Face
First-run notice
On first launch or when loading a model for the first time, setup may take several minutes while Python, dependencies, and model files are downloaded and initialized. This is expected behavior.
No manual Python setup is required.
Device Support
Supports CUDA and ROCM based devices
Important
This version will overwrite your current Settings.json, if you have added environments or models you may want to backup this file first
v0.3.7 - Gallery & Preview
Gallery & Preview
Added image/video gallery and preview popup
Features
- Gallery & Preview
- Improved pipeline loading
- Prompt caching
- Other bug-fixes
Full Changelog: v0.3.6...v0.3.7
Installation
1. Installer Version
- Download and run Diffuse_v0.3.7.exe
- Follow the on-screen instructions
2. Portable Version
-
Download and extract Diffuse_v0.3.7.zip
A fast SSD with plenty of free space is recommended, as model downloads can be large. -
Run Diffuse.exe
Diffuse will automatically:
- Install an isolated portable Python runtime
- Create the required virtual environment
- Download the selected model from Hugging Face
First-run notice
On first launch or when loading a model for the first time, setup may take several minutes while Python, dependencies, and model files are downloaded and initialized. This is expected behavior.
No manual Python setup is required.
Device Support
Supports CUDA and ROCM based devices
Important
This version will overwrite your current Settings.json, if you have added environments or models you may want to backup this file first
v0.3.6 - Upscale & Interpolation Views
Upscale & Interpolation Views
Support for image & video upscaling, interpolation and feature extraction
Features
- Upscale & Interpolation Views
- Existing Huggingface cache detection
- Diffuse installer (optional)
- Stability and other bug-fixes
- End of
Alphastage
Installation
1. Installer Version
- Download and run Diffuse_v0.3.6.exe
- Follow the on-screen instructions
2. Portable Version
-
Download and extract Diffuse_v0.3.6.zip
A fast SSD with plenty of free space is recommended, as model downloads can be large. -
Run Diffuse.exe
Diffuse will automatically:
- Install an isolated portable Python runtime
- Create the required virtual environment
- Download the selected model from Hugging Face
First-run notice
On first launch or when loading a model for the first time, setup may take several minutes while Python, dependencies, and model files are downloaded and initialized. This is expected behavior.
No manual Python setup is required.
Device Support
Supports CUDA and ROCM based devices
v0.3.5 - Environment Manager
Environment Manager
Support for multiple python virtual environments, Vendor, Device and Pipeline based environments
Features
- Isolated python process
- Custom Python virtual environments
- Environment variable support
Installation
-
Download and extract Diffuse_v0.3.5.zip
A fast SSD with plenty of free space is recommended, as model downloads can be large. -
Run Diffuse.exe
-
Load a model
Diffuse will automatically:- Install an isolated portable Python runtime
- Create the required virtual environment
- Download the selected model from Hugging Face
First-run notice
On first launch or when loading a model for the first time, setup may take several minutes while Python, dependencies, and model files are downloaded and initialized. This is expected behavior.
No manual Python setup is required.
Device Support
Supports CUDA and ROCM based devices
v0.3.2 - Diffuse - Proof Of Concept
Diffuse - Proof Of Concept
Diffuse is a Windows desktop UI for Huggingface Diffusers. It integrates directly with Python using the Python C API via CSnakes, enabling high-performance interop between .NET and Python for running diffusion models.
Features
- Automatic installation of isolated portable Python
- Device-specific Python virtual environments
- Automatic model downloads from Huggingface repositories
Supported Pipelines
- Z-Image: ZImagePipeline, ZImageImg2ImgPipeline
- Qwen Image: QwenImagePipeline, QwenImageImg2ImgPipeline, QwenImageEditPlusPipeline
- FLUX.1: FluxPipeline, FluxImg2ImgPipeline, FluxKontextPipeline, FluxControlNetPipeline
- FLUX.2: Flux2Pipeline
- Chroma: ChromaPipeline, ChromaImg2ImgPipeline
- LTX-Video: LTXPipeline, LTXImageToVideoPipeline
- Wan Video: WanPipeline, WanImageToVideoPipeline
- CogVideoX: CogVideoXPipeline, CogVideoXImageToVideoPipeline, CogVideoXVideoToVideoPipeline
- Kandinsky5: Kandinsky5T2IPipeline, Kandinsky5I2IPipeline, Kandinsky5T2VPipeline, Kandinsky5I2VPipeline
- StableDiffusionXL: StableDiffusionXLPipeline, StableDiffusionXLImg2ImgPipeline, StableDiffusionXLControlNetPipeline, StableDiffusionXLControlNetImg2ImgPipeline
Installation
-
Download and extract Diffuse_v0.3.2_CUDA.zip
A fast SSD with plenty of free space is recommended, as model downloads can be large. -
Run Diffuse.exe
-
Load a model
Diffuse will automatically:- Install an isolated portable Python runtime
- Create the required virtual environment
- Download the selected model from Hugging Face
First-run notice
On first launch or when loading a model for the first time, setup may take several minutes while Python, dependencies, and model files are downloaded and initialized. This is expected behavior.
No manual Python setup is required.
Device Support
Supports CUDA based devices with support for ROCM based devices in active development
