Trt-engine is 2x slower than the onnx model

Description

I tried to inference the sam2-decoder with trt-engine, but, as you can see from the screenshots below, it is much slower than the inference with onnx model.
TRT

ORT

For better investigating, the onnx model is in attachment.

Any clue would be appreciated, thanks in advance!

Environment

TensorRT Version: 10.3.0

NVIDIA GPU: Jetson Orin 64GB

NVIDIA Driver Version: 540.4.0

CUDA Version: 12.6

CUDNN Version: 9.3.0

Operating System:

Python Version (if applicable): 3.10.12

Tensorflow Version (if applicable):

PyTorch Version (if applicable):

Baremetal or Container (if so, version):

Relevant Files

Model link:
Model for Debugging

Steps To Reproduce

Commands or scripts:

Have you tried the latest release?:

Can this model run on other frameworks? For example run ONNX model with ONNXRuntime (polygraphy run <model.onnx> --onnxrt):

The payload names only the attached ONNX model and the sam2-decoder; no repository file, test, or runnable command is provided. Start by reproducing the TensorRT and ONNXRuntime comparison in the listed environment, then document the cause and verify that TensorRT performance is comparable under the same conditions.