sdcb/PaddleSharp

.NET/C# binding for Baidu paddle inference library and PaddleOCR

What it solves

PaddleSharp is a .NET wrapper for the PaddleInference C API, allowing C# developers to integrate powerful AI models from the PaddlePaddle ecosystem into their applications. It bridges the gap between the C++ based inference engine and the .NET environment, providing native support for Windows, Linux, and macOS.

How it works

The project provides a set of .NET bindings for the Paddle Inference C API. It offers various runtime packages tailored to specific hardware acceleration options, including:

  • CPU Acceleration: Support for MKL (recommended for most users), OpenBLAS, and OpenVINO.
  • GPU Acceleration: Extensive support for NVIDIA CUDA (versions 11.8, 12.6, and 12.9) across a wide range of GPU architectures (GTX 10 series through RTX 50 series).
  • ONNX Runtime: Integration with onnxruntime for efficient inference.

Who it’s for

C#/.NET developers who want to use PaddlePaddle models for tasks like OCR, object detection, and NLP without needing to write C++ code.

Highlights

  • Comprehensive AI Capabilities: Built-in support for PaddleOCR (14 languages, table recognition), PaddleDetection (PPYolo, PicoDet), Rotation Detection, and PaddleNLP (Chinese segmenter).
  • Cross-Platform Support: Works on Windows (x64), Linux (Ubuntu-22.04 x64/ARM64), and macOS (x64/ARM64).
  • Flexible Hardware Acceleration: Supports a wide array of NVIDIA GPUs and CPU-based acceleration libraries.
  • Model Export: Includes Paddle2Onnx to allow users to export models to ONNX format using C#.

Related

  • Project
    writer/writer-framework

    An open-source framework for building AI applications that combines a visual drag-and-drop UI editor with a Python-based backend.

  • Project
    wannabespace/conar

    Tamery is an AI-powered open-source database manager that helps users securely manage connections and use AI to write and optimize SQL queries.

  • Project
    PolyArch/humanize

    A Claude Code plugin that implements an iterative development loop with independent AI review to ensure high-quality, bug-free code.

  • Project
    sauravpanda/BrowserAI

    A JavaScript SDK that enables running production-ready LLMs and multimodal AI models directly in the browser using WebGPU acceleration for privacy and zero server costs.