v0.3.0 - Python Pipelines

Python Pipelines

You can now run Diffusers Python pipelines directly from C#, with everything managed for you.

• Automatic Python Setup
  Downloads and installs an isolated Python runtime — no system Python required.

• Device-Specific Virtual Environments
  Creates a dedicated virtual environment tailored to the selected device.

• Hugging Face Model Downloads
  Automatically downloads Diffusers models from Hugging Face.

• LoRA Support
  Load and apply LoRA models in supported pipelines.

• CUDA Support
  Support for CUDA devices with support for ROCM in development

• Window Demo App
  Basic Window PythonDemo App available for download below

Supported Diffusers Pipelines

Chroma

• ChromaPipeline
• ChromaImg2ImgPipeline

Qwen Image

• QwenImagePipeline
• QwenImageImg2ImgPipeline
• QwenImageEditPipeline

Wan Video

• WanPipeline
• WanImageToVideoPipeline

Z-Image

• ZImagePipeline
• ZImageImg2ImgPipeline

---

Basic Example

Python Virtual Environment

The PythonManager can be used to download and install Python and create virtual environments

// Virtual Environment Config
var serverConfig = new EnvironmentConfig
{
    Environment = "default-cuda",
    Directory = "PythonRuntime",
    Requirements =
    [
        "--extra-index-url https://download.pytorch.org/whl/cu118",
        "torch==2.7.0+cu118",
        "typing",
        "wheel",
        "transformers",
        "accelerate",
        "diffusers",
        "protobuf",
        "sentencepiece",
        "pillow",
        "ftfy",
        "scipy",
        "peft"
    ]
};

// PythonManager
var pythonManager = new PythonManager(serverConfig);

// Create/Load Virtual Environment
await pythonManager .CreateEnvironmentAsync(PipelineProgress.ConsoleCallback);

Python Pipelines

Once you have created a virtual environment you can now load a pipeline

// Pipeline Config
var pipelineConfig = new PipelineConfig
{
    Path = "Qwen/Qwen-Image-Edit",
    Pipeline = "QwenImagePipeline",
    ProcessType = ProcessType.ImageEdit,
    IsFullOffloadEnabled = true,
    DataType = DataType.Bfloat16
};

// Create Pipeline Proxy
using (var pythonPipeline = new PythonPipeline(pipelineConfig, PipelineProgress.ConsoleCallback))
{
    // Download/Load Model
    await pythonPipeline.LoadAsync();

    // Generate Option
    var options = new PipelineOptions
    {
        Prompt = "Yarn art style",
        Steps = 30,
        Width = 1024,
        Height = 1024,
        GuidanceScale = 4f,
        Scheduler = SchedulerType.FlowMatchEulerDiscrete,
        ImageInput = new ImageInput("Image.png")
    };

    // Generate
    var response = await pythonPipeline.GenerateAsync(options);

    // Save Image
    await response
        .AsImageTensor()
        .SaveAsync("Result.png");
}

Full Changelog: v0.2.2...v0.3.0

v0.2.2 - .NET10 Support

.NET 10 Support

This update adds .NET 10 as a build target, demo applications updated

What’s New

• Support for .NET 10
• Additional Supertonic TTS voices

Full Changelog: v0.1.84...v0.2.2

v0.1.84 - Supertonic Demo App

Supertonic TTS Support Added

This update brings Supertonic Text-to-Speech support to the TextDemo application
You can now generate high-quality speech directly inside the app with no extra setup required.

What’s New

• Supertonic TTS Integration: High-quality, expressive speech synthesis with fast inference.
• Real-time Parameter Controls: Adjust speed, steps, and style directly from the UI.
• Export to WAV: Save generated audio instantly.

Full Changelog: v0.1.81...v0.1.84

v0.1.81 - Supertonic Pipeline

What's Changed

• Prompt cache by @saddam213 in #4
• Supertonic TTS Pipeline by @saddam213 in #7

Full Changelog: v0.1.39...v0.1.81

v0.1.39 - AMD Nitro Pipeline

What's Changed

• AMD Nitro Pipeline
• Llama-3.2-1B pipeline
• ImageTensor GetPixel/SetPixel support
• Video background/foreground masking,removal

New Contributors

• @jdluzen for the support and contribution to #2

Full Changelog: v0.1.18...v0.1.39

v0.1.18 - OpenPose Support

OpenPose Extractor

Features

• Image pose extraction
• Video pose extraction
• Windows demo for image and video

Full Changelog: v0.1.17...v0.1.18

Extractor Demo App

TensorStack Extractor Demo App

The first release of the TensorStack Extractor Demo, a WPF application showcasing the capabilities of TensorStack.Extractors project.

Features

• Simple WPF interface for image and video
• Depth, SoftEdge, Line-Art support
• Background/Foreground Removal
• Background/Foreground Masking

Upscaler Demo App

TensorStack Upscaler Demo App

The first release of the TensorStack Upscaler Demo, a WPF application showcasing the capabilities of TensorStack.Upscaler.

This demo allows users to upscale images and videos using the TensorStack backend, with optional RIFE frame interpolation for smoother playback.

Features

• Simple WPF interface for image and video upscaling
• Comparison view
• Support for RIFE frame interpolation

Image/Video Upscale & Interpolation

TensorStack.Upscaler

The first official release of TensorStack.Upscaler, part of the TensorStack suite.
This library provides high-performance image and video upscaling using ONNX Runtime and supports both buffered inputs (e.g. file-based) and streaming sources (e.g. camera, pipeline, or frame feeds).

Key Features

• Image Upscaling – Enhance image resolution with minimal loss in quality.
• Video Upscaling – Process video frames from file or stream sources with synchronized output.
• RIFE Frame Interpolation – Integrated RIFE model for smooth motion and higher effective frame rates.
• Flexible Input Handling – Supports buffered and real-time processing pipelines.
• TensorStack Integration – Built to work seamlessly alongside other TensorStack components like TensorStack.Image and TensorStack.Video.

Technical Notes

• Built on ONNX Runtime with multi-provider support (CPU, CUDA, DirectML).
• Compatible with TensorStack.Image & TensorStack.Video tensors.

Known Limitations

• Interpolation and upscale models currently run sequentially per frame (parallel batching planned).
• Limited metadata passthrough for streaming sources.

Docs & Examples

TensorStack.Upscaler Readme

Nuget Packages

https://www.nuget.org/packages/TensorStack.Upscaler
https://www.nuget.org/packages/TensorStack.Common
https://www.nuget.org/packages/TensorStack.Providers.CPU
https://www.nuget.org/packages/TensorStack.Providers.DML
https://www.nuget.org/packages/TensorStack.Providers.CUDA
https://www.nuget.org/packages/TensorStack.Image
https://www.nuget.org/packages/TensorStack.Image.Bitmap
https://www.nuget.org/packages/TensorStack.Image.BitmapImage
https://www.nuget.org/packages/TensorStack.Image.ImageSharp
https://www.nuget.org/packages/TensorStack.Video
https://www.nuget.org/packages/TensorStack.Video.Windows