OnnxExecutionProvider
Enums · AiDotNet.Onnx
Specifies the execution provider (hardware accelerator) for ONNX model inference.
For Beginners
Think of execution providers as different engines:
- CPU: Works everywhere, slowest but most compatible
- CUDA: NVIDIA GPUs, much faster than CPU
- TensorRT: NVIDIA GPUs with extra optimizations, fastest for NVIDIA
- DirectML: Windows GPUs (AMD, Intel, NVIDIA), good cross-vendor support
- CoreML: Apple Silicon (M1/M2/M3), fastest on Mac
How It Works
Execution providers allow ONNX models to run on different hardware accelerators. The order of fallback is typically: CUDA/TensorRT → DirectML → CPU.
Fields
| Field | Summary |
|---|---|
OnnxExecutionProvider Auto |
Automatically select the best available provider. |
OnnxExecutionProvider CoreML |
Apple CoreML execution provider for Apple Silicon. |
OnnxExecutionProvider Cpu |
CPU execution provider (default, always available). |
OnnxExecutionProvider Cuda |
NVIDIA CUDA execution provider for GPU acceleration. |
OnnxExecutionProvider DirectML |
DirectML execution provider for Windows GPU acceleration. |
OnnxExecutionProvider NNAPI |
NNAPI execution provider for Android devices. |
OnnxExecutionProvider OpenVINO |
OpenVINO execution provider for Intel hardware. |
OnnxExecutionProvider ROCm |
ROCm execution provider for AMD GPUs. |
OnnxExecutionProvider TensorRT |
NVIDIA TensorRT execution provider for optimized GPU inference. |
int value__ |