onnxruntime

ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator

Run the following Vcpkg command to install the port.

vcpkg install onnxruntime

Usage details are not available for this port.

• onnxruntime[cuda]≥v1.23.2#0

  (x64 & windows & !static) | (x64 & linux)

  Build with CUDA support

  Feature Dependencies

  • cuda

    All
  • cudnn

    All
  • cudnn-frontend

    All
  • nvidia-cutlass

    All
• onnxruntime[framework]≥v1.23.2#0

  osx | ios

  Build a macOS/iOS framework, Objective-C library

• onnxruntime[openvino]≥v1.23.2#0

  !(osx | ios | android | emscripten)

  Build with OpenVINO support

  Feature Dependencies

  • openvino

    All

    • openvino[cpu]
      All

    • openvino[gpu]
      All

    • openvino[onnx]
      All
• onnxruntime[tensorrt]≥v1.23.2#0

  All

  Build with TensorRT support

  Feature Dependencies

  • onnxruntime

    All

    • onnxruntime[cuda]
      All

• abseil≥v20250512.1#0

  All
• boost-config≥v1.82.0#0

  All
• boost-mp11≥v1.82.0#0

  All
• cpuinfo

  All
• cxxopts

  All
• date

  All
• dlpack

  All
• eigen3

  All
• flatbuffers≥v23.5.26#0

  All
• flatbuffers≥v23.5.26#0

  All
• ms-gsl

  All
• nlohmann-json

  All
• onnx≥v1.19.0#0

  All
• optional-lite

  All
• protobuf≥v3.21.12#0

  All
• protobuf≥v3.21.12#0

  All
• re2

  All
• safeint

  All
• utf8-range

  All
• vcpkg-cmake

  All
• vcpkg-cmake-config

  All
• wil

  windows