TensorRT models not compatible with on another device

[fadoughou]

I have converted an onnx model to TRT 8.5.2 using trtexec.
the conversion occurred on a server and could run inference on NVIDIA GPU.
However, when deploying the model on Jetson AGX orin device, it shows error and cannot run.
Is it necessary to convert a model on the same device where inference will run?

[spolisetty]

@fadoughou

Is it necessary to convert a model on the same device where inference will run?

Yes. For Jetson we need to build the TensorRT engine on the same device.
For more information, please refer to the following similar post:

Convert ONNX to TensorRT and run on another device

Hi, Please refer to the following docs, which may help you. By default, TensorRT engines are only compatible with the type of device where they were built. With build-time configuration, engines can be built that are compatible with other types of devices. Currently, hardware compatibility is supported only for Ampere and later device architectures and is not supported on NVIDIA DRIVE OS or JetPack. Thank you.

Thank you.