Skip to main content
Ctrl+K
onnx-light-cpu documentation - Home onnx-light-cpu documentation - Home
  • Getting Started
  • Design
  • ByOp
  • API reference
  • Examples
  • Next Steps
  • GitHub
  • Getting Started
  • Design
  • ByOp
  • API reference
  • Examples
  • Next Steps
  • GitHub

Section Navigation

  • Standalone C++ inference
  • Kernel examples
    • Runtime SIMD detection
    • Run com.microsoft custom operators
  • Benchmarks
    • Benchmark custom operators against ONNX Runtime
    • Benchmark Exp and Log parallel scheduling
    • Benchmark TreeEnsemble scheduling scenarios
    • Benchmark Gemm: numpy vs onnxruntime vs onnx-light vs onnx-light-cpu
    • Benchmark Abs: onnxruntime vs onnx-light + onnx-light-cpu
    • Benchmark Gemm: float32 vs float16 vs bfloat16 across kernel code paths
    • Benchmark backend test cases against ONNX Runtime
  • Processor performance
    • Processor performance profile: memory, compute, and Roofline
  • Examples

Examples#

Runnable examples showing how to use the SIMD-accelerated CPU kernels provided by onnx-light-cpu from Python and C++.

  • Standalone C++ inference
  • Kernel examples
  • Benchmarks
  • Processor performance

previous

Runtime and SIMD dispatch

next

Standalone C++ inference
Show Source

Last updated on Sep 07, 2026.