onnxruntime

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

Latest version: 1.22.0

Add to xmake.lua

add_requires("onnxruntime 1.22.0")

Learn how to add and link packages with Xmake →

Information

License
MIT
Recipe updated
2026-07-16T14:16:04.000Z

Platforms

  • linux: x86_64, arm64
  • macosx: x86_64, arm64
  • windows: x86, x64, arm, arm64, arm64ec

Options

cuda_version
Specify which CUDA version to use for GPU support · Type: string · Default: 11
gpu
Enable GPU support on windows|x64 and linux|x86_64 · Type: boolean · Default: false
shared
Download shared binaries. · Type: boolean · Default: true

Versions

  • 1.11.1
  • 1.14.1
  • 1.15.1
  • 1.16.1
  • 1.17.0
  • 1.17.1
  • 1.19.2
  • 1.22.0

Dependencies

No dependencies declared.

Install with xrepo

Install the package independently of a project and inspect its available configuration.

install

xrepo install "onnxruntime 1.22.0"

show info

xrepo info "onnxruntime"

Link a library to your target

Declare the requirement, then attach the library to the target that uses it.

xmake.lua

add_requires("onnxruntime 1.22.0")

target("demo")
    set_kind("binary")
    add_files("src/*.cpp")
    add_packages("onnxruntime")

Open a package environment

Open a shell with the environment exported by this package. Use the executable names documented by the package.

enter shell

xrepo env -b "onnxruntime 1.22.0" shell