Docker Image for JetPack 7.1

Robotics & Edge Computing Jetson Systems Jetson Thor
1

JetPack 7.1 is released today, When will the docker image available on ngc? Need to run it on Thor. Thanks.

4

Hi,

What kind of container do you need?
There are several containers (ex., vLLM) that can run on JetPack 7.1.

Thanks.

6

Same question here. None of the base images seem to support Jetpack 7.1 / l4t r38 yet.

7

On this webpage click Tags https://catalog.ngc.nvidia.com/orgs/nvidia/containers/cuda

If still on cuda-13.0 scroll down a bunch to see the cuda-13.0.2 Ubuntu 24.04 images. Pick the one with the features you need. It is arm64 and for me works on Thor r38.4.0 Jetpack 7.1. It also works well as a base image in Dockerfile.

run container with --runtime nvidia

On this webpage for Tensorrt docker image pick the ubuntu 24.04 with the version of Tensorrt you want.https://docs.nvidia.com/deeplearning/frameworks/container-release-notes/

Then go here to get the path. https://catalog.ngc.nvidia.com/orgs/nvidia/containers/tensorrt

root@chithor:/# cat /etc/nv_tegra_release 
# R38 (release), REVISION: 4.0, GCID: 43443517, BOARD: generic, EABI: aarch64, DATE: Wed Dec 31 00:15:19 UTC 2025
# KERNEL_VARIANT: oot
TARGET_USERSPACE_LIB_DIR=nvidia
TARGET_USERSPACE_LIB_DIR_PATH=usr/lib/aarch64-linux-gnu/nvidia
INSTALL_TYPE=openrm

root@chithor:/# cat /etc/lsb-release 
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=24.04
DISTRIB_CODENAME=noble
DISTRIB_DESCRIPTION="Ubuntu 24.04.3 LTS"

ngc registry image info nvcr.io/nvidia/cuda:13.0.2-runtime-ubuntu24.04
------------------------------------------------------------------------------------
 Image Information
   Name: nvidia/cuda:13.0.2-runtime-ubuntu24.04
   Architecture: arm64
     Image Size: 1.65 GB
     Digest: sha256:bbfc230dfd4c3bb2eadacdc22cb7e2ea3453e93bb3ee40e241357eb0bcd5c6a6
   Schema Version: 1
   Signed?: True
   Last Updated: Nov 06, 2025
8

Hi,

Our internal team is discussing this request.
Will share more information with you later.

Thanks.

9

Hi,

There are some DLFW containers available.
Please use them instead:
vLLM: https://catalog.ngc.nvidia.com/orgs/nvidia/containers/vllm?version=26.01-py3
SGLang: SGLang | NVIDIA NGC

Thanks.