Convert From TensorRT to OpenVINO

TensorRT and OpenVINO are are model optimization and inference acceleration toolkits from NVIDIA and Intel, respectively.

ToDo…

Comparison Summary

Feature	TensorRT	OpenVINO
Vendor	NVIDIA	Intel
Hardware target	NVIDIA GPUs	Intel CPUs, iGPUs, GPUs, NPUs, VPUs, FPGAs
Model input formats	ONNX, TensorFlow, PyTorch (via ONNX)	TensorFlow, ONNX, PyTorch (via ONNX)
Optimizations	Layer fusion, precision calibration, kernel tuning	Graph simplification, quantization, fusion
Output	GPU “engine” binary	OpenVINO IR model
Precision modes	FP32, FP16, INT8	FP32, FP16, INT8
Serving integration	Triton Inference Server	OpenVINO Model Server

Conversion

ToDo…

Migration Example: Lorem Ipsum

Quis tempor aliquip et enim aliqua. Aliquip cupidatat laboris ipsum aute eiusmod. Aliqua est adipisicing anim consequat minim nostrud aute aute pariatur dolore eiusmod qui laborum proident. Officia nisi et esse excepteur magna elit pariatur. Dolore consectetur ipsum occaecat eu. Mollit laboris ex amet commodo eiusmod mollit. Eiusmod consectetur sit laborum elit officia aute exercitation adipisicing tempor.