# TensorRT installation issue on AGX ORIN

**URL:** <https://forums.developer.nvidia.com/t/tensorrt-installation-issue-on-agx-orin/250019>\
**Category:** TensorRT\
**Tags:** tensorrt\
**Created:** [April 15, 2023, 4:29am UTC](https://forums.developer.nvidia.com/t/tensorrt-installation-issue-on-agx-orin/250019 "2023-04-15T04:29:42Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![darshana](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@darshana](https://forums.developer.nvidia.com/u/darshana)\
**Post date:** [April 15, 2023, 4:29am UTC](https://forums.developer.nvidia.com/t/tensorrt-installation-issue-on-agx-orin/250019/1 "2023-04-15T04:29:42Z")

</div>

The default TensorRT version which is installed on ORIN with Jetpack 5.1/5.0 is 8.5( with ubuntu 20.04 , python 3.8) . When the OS is upgraded to Ubuntu 22.04, then the default python Version has 3.10.

- After setting the default python path to 3.8, we are facing an error related to TensorRT (both on python3.8, python3.10) though the package files are present in user site-packages.

![trt](https://global.discourse-cdn.com/nvidia/original/3X/9/2/926c96043b384b75037403ef7a9f99e360d32bdb.png)

What would be the suitable solution for this bug ?

- We would want to have a tensorrt version which is both compatible to Python version 3.8 and python 3.10 on CUDA 11.4, Ubuntu 22.04. Or any of the compatible version of trt which runs on either mentioned python versions (on CUDA 11.4, Ubuntu 22.04)

---

<div class="post-metadata">

**Author:** ![AakankshaS](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/aakankshas/32/14047_2.png) [@AakankshaS](https://forums.developer.nvidia.com/u/AakankshaS)\
**Post date:** [April 15, 2023, 4:37am UTC](https://forums.developer.nvidia.com/t/tensorrt-installation-issue-on-agx-orin/250019/2 "2023-04-15T04:37:48Z")

</div>

Hi,  
Please refer to the installation steps from the below link if in case you are missing on anything

> **[Installation Guide :: NVIDIA Deep Learning TensorRT Documentation](https://docs.nvidia.com/deeplearning/tensorrt/install-guide/index.html)**
>
> This NVIDIA TensorRT 8.6.0 Early Access (EA) Installation Guide provides the installation requirements, a list of what is included in the TensorRT package, and step-by-step instructions for installing TensorRT.

Also, we suggest you to use TRT NGC containers to avoid any system dependency related issues.

> **[TensorRT | NVIDIA NGC](https://catalog.ngc.nvidia.com/orgs/nvidia/containers/tensorrt)**
>
> NVIDIA TensorRT is a C++ library that facilitates high-performance inference on NVIDIA graphics processing units (GPUs). TensorRT takes a trained network and produces a highly optimized runtime engine that performs inference for that network.

Thanks!