ONNX — why use it?

medium

Answer

$Open Neural Network Exchange: standard format for cross-framework interop.$
$Train in PyTorch \to export ONNX \to run in ONNX Runtime (C++, C#, Java, JS).$
$Benefits: (1) framework-agnostic deployment.$
$(2) ONNX Runtime is optimized (better than raw PyTorch inference).$
$(3) edge deployment (mobile, embedded).$
$(4) hardware-specific execution providers (CUDA, TensorRT, OpenVINO, DirectML).$
$Downside: not all ops supported; custom ops require conversion.$
$Standard for edge / cross-platform.$

Check yourself — multiple choice

Random
Open format for cross-framework interop: PyTorch export → ONNX Runtime (C++/C#/Java/JS); optimized runtime + hardware providers (CUDA/TRT/OpenVINO/DirectML); edge + cross-platform standard
Same as PyTorch
Not real

ONNX: cross-framework format; hardware providers; edge + cross-platform.

#serving#infrastructure

215 interview questions in this topic.