# Docker and NVIDIA Docker **URL:** https://forums.developer.nvidia.com/c/accelerated-computing/nvidia-gpu-cloud-ngc-users/docker-and-nvidia-docker/33.md [Latest](https://forums.developer.nvidia.com/latest.md) · [Categories](https://forums.developer.nvidia.com/categories.md) · [Tags](https://forums.developer.nvidia.com/tags.md) --- ## [Docker Image for CUDA 13.3.1?](https://forums.developer.nvidia.com/t/docker-image-for-cuda-13-3-1/378207)
**Author:** [@03yfli](https://forums.developer.nvidia.com/u/03yfli)\ **Replies:** 0\ **Last updated:** [July 26, 2026, 9:00pm UTC](https://forums.developer.nvidia.com/t/docker-image-for-cuda-13-3-1/378207 "2026-07-26T21:00:35Z")
Hello Nvidia Team, CUDA 13.3.1 was released about a month ago, but the docker images are still not available at nvidia/cuda - Docker Image. May I ask when they will be released? We need that version as it contains a sp… --- ## [NVIDIA Container toolkit in Ubuntu 22.04 LTS - NO\_PUBKEY DDCAE044F796ECB0](https://forums.developer.nvidia.com/t/nvidia-container-toolkit-in-ubuntu-22-04-lts-no-pubkey-ddcae044f796ecb0/337001)
**Author:** [@sajith.arimbra](https://forums.developer.nvidia.com/u/sajith.arimbra)\ **Replies:** 1\ **Last updated:** [June 25, 2026, 11:24am UTC](https://forums.developer.nvidia.com/t/nvidia-container-toolkit-in-ubuntu-22-04-lts-no-pubkey-ddcae044f796ecb0/337001 "2026-06-25T11:24:36Z")
I am trying to install NVIDIA container toolkit in Ubuntu 22.04 LTS. I have added the apt repo and related keyring too. Below is my apt repo list deb \[signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg\]… --- ## [Triton Inference Server Support Matrix lists incorrect PyTorch version for release 26.05](https://forums.developer.nvidia.com/t/triton-inference-server-support-matrix-lists-incorrect-pytorch-version-for-release-26-05/373946)
**Author:** [@alejandrocarmonamartinez6](https://forums.developer.nvidia.com/u/alejandrocarmonamartinez6)\ **Replies:** 0\ **Last updated:** [June 20, 2026, 1:08pm UTC](https://forums.developer.nvidia.com/t/triton-inference-server-support-matrix-lists-incorrect-pytorch-version-for-release-26-05/373946 "2026-06-20T13:08:36Z")
Hello, My name’s Alejandro, I’m a first-year PhD student in CS and this is my first post. Description I was using the 26.05 docker image for the Triton Inference Server and according to the Frameworks Support Matrix, t… --- ## [Docker container image for recent vLLM release that enables GGUF loading](https://forums.developer.nvidia.com/t/docker-container-image-for-recent-vllm-release-that-enables-gguf-loading/349106)
**Author:** [@nhira](https://forums.developer.nvidia.com/u/nhira)\ **Replies:** 4\ **Last updated:** [April 14, 2026, 5:58pm UTC](https://forums.developer.nvidia.com/t/docker-container-image-for-recent-vllm-release-that-enables-gguf-loading/349106 "2026-04-14T17:58:21Z")
As explained in this post, the community would appreciate a more recent NGC vLLM image. The current image (nvcr.io/nvidia/vllm:25.09-py3) does not support GGUF loading. The workaround is to compile against the 25.09 ba… --- ## [Cannot pull nvcr.io/nvidia/nre containers – "Payment Required" (NuRec / NRE)](https://forums.developer.nvidia.com/t/cannot-pull-nvcr-io-nvidia-nre-containers-payment-required-nurec-nre/365098)
**Author:** [@ddvtiras](https://forums.developer.nvidia.com/u/ddvtiras)\ **Replies:** 1\ **Last updated:** [March 30, 2026, 2:07pm UTC](https://forums.developer.nvidia.com/t/cannot-pull-nvcr-io-nvidia-nre-containers-payment-required-nurec-nre/365098 "2026-03-30T14:07:50Z")
Hello, I am currently integrating NVIDIA NuRec (Neural Reconstruction Engine) into our 3D reconstruction pipeline. docker login nvcr.io → Login Succeeded Context: We are evaluating NVIDIA NuRec for a production-grade… --- ## [About a suggestion from my side for better growth purpose](https://forums.developer.nvidia.com/t/about-a-suggestion-from-my-side-for-better-growth-purpose/364248)
**Author:** [@princeradadiya92](https://forums.developer.nvidia.com/u/princeradadiya92)\ **Replies:** 0\ **Last updated:** [March 21, 2026, 10:20am UTC](https://forums.developer.nvidia.com/t/about-a-suggestion-from-my-side-for-better-growth-purpose/364248 "2026-03-21T10:20:23Z")
Just a suggestion from my side just attended a event from collabnix x Nvidia GTC watchparty where all things were good also it was having sessions regarding jetson nano projects and about nemoclaw and nvidia has shipped … --- ## [NVAIE change organization key associate to license](https://forums.developer.nvidia.com/t/nvaie-change-organization-key-associate-to-license/362637)
**Author:** [@moridaira.kenji](https://forums.developer.nvidia.com/u/moridaira.kenji)\ **Replies:** 0\ **Last updated:** [March 6, 2026, 10:38am UTC](https://forums.developer.nvidia.com/t/nvaie-change-organization-key-associate-to-license/362637 "2026-03-06T10:38:43Z")
Now is using NVIDIA AI Enterprise Essentials(Evaluation), but we associate individual personal key not the organization key to the entitlement license. Can NVIDIA change the individual key to organization key and asso… --- ## [GPU Access Issue: CUDA Only Runs via Sudo inside Docker Dev Container (Orin NX / JP 6.1)](https://forums.developer.nvidia.com/t/gpu-access-issue-cuda-only-runs-via-sudo-inside-docker-dev-container-orin-nx-jp-6-1/361027)
**Author:** [@anirudhaduttagupta2004](https://forums.developer.nvidia.com/u/anirudhaduttagupta2004)\ **Replies:** 0\ **Last updated:** [February 19, 2026, 9:17am UTC](https://forums.developer.nvidia.com/t/gpu-access-issue-cuda-only-runs-via-sudo-inside-docker-dev-container-orin-nx-jp-6-1/361027 "2026-02-19T09:17:37Z")
Hardware: Jetson Orin NX 16GB JetPack Version: r36.4.0 (L4T) Base Image: nvcr.io/nvidia/l4t-jetpack:r36.4.0 Hi everyone, I’m setting up a development environment on a Jetson Orin NX using VS Code Dev Containers and h… --- ## [TensorRT-LLM PyTorch Memory Fraction Automatically Limited to 0.14 Despite KV Cache Configuration](https://forums.developer.nvidia.com/t/tensorrt-llm-pytorch-memory-fraction-automatically-limited-to-0-14-despite-kv-cache-configuration/354807)
**Author:** [@sh.ahn1](https://forums.developer.nvidia.com/u/sh.ahn1)\ **Replies:** 0\ **Last updated:** [December 16, 2025, 6:57am UTC](https://forums.developer.nvidia.com/t/tensorrt-llm-pytorch-memory-fraction-automatically-limited-to-0-14-despite-kv-cache-configuration/354807 "2025-12-16T06:57:39Z")
When serving the GPT-OSS-20B model using TensorRT-LLM’s trtllm-serve command, the PyTorch memory fraction is automatically limited to 0.14 (14%) despite explicitly configuring KV cache memory through both YAML configurat… --- ## [Sharing GPU between Docker containers](https://forums.developer.nvidia.com/t/sharing-gpu-between-docker-containers/354197)
**Author:** [@avinashkushwah31](https://forums.developer.nvidia.com/u/avinashkushwah31)\ **Replies:** 0\ **Last updated:** [December 10, 2025, 2:24pm UTC](https://forums.developer.nvidia.com/t/sharing-gpu-between-docker-containers/354197 "2025-12-10T14:24:34Z")
I am running multiple Docker containers using the following command, which exposes all GPU devices to each container: sudo docker run --rm --runtime=nvidia --gpus all ubuntu nvidia-smi Scenario: Container C1 runs P… --- ## [Expected behavior with NVIDIA GPU Operator and time-slicing on server with multiple gpus](https://forums.developer.nvidia.com/t/expected-behavior-with-nvidia-gpu-operator-and-time-slicing-on-server-with-multiple-gpus/353675)
**Author:** [@scott\_selberg](https://forums.developer.nvidia.com/u/scott_selberg)\ **Replies:** 0\ **Last updated:** [December 5, 2025, 1:34pm UTC](https://forums.developer.nvidia.com/t/expected-behavior-with-nvidia-gpu-operator-and-time-slicing-on-server-with-multiple-gpus/353675 "2025-12-05T13:34:13Z")
Hi, I have a single server with 4 identical gpus connected to OpenShift. Multiple teams plan to deploy LLM enabled apps on the server. We know that these apps will sit idle till a request comes in, so time-slicing see… --- ## [/etc/nv\_tegra\_release missing from docker images](https://forums.developer.nvidia.com/t/etc-nv-tegra-release-missing-from-docker-images/353453)
**Author:** [@pierce](https://forums.developer.nvidia.com/u/pierce)\ **Replies:** 3\ **Last updated:** [December 3, 2025, 10:28pm UTC](https://forums.developer.nvidia.com/t/etc-nv-tegra-release-missing-from-docker-images/353453 "2025-12-03T22:28:45Z")
Hi There! I am using the l4t base docker image from here: NVIDIA L4T Base | NVIDIA NGC One thing I am finding is the /etc/nv\_tegra\_release file is missing when running the image: Is there any way to determine the L4… --- ## [Nvcr.io is down](https://forums.developer.nvidia.com/t/nvcr-io-is-down/352247)
**Author:** [@nektarios](https://forums.developer.nvidia.com/u/nektarios)\ **Replies:** 0\ **Last updated:** [November 22, 2025, 3:16am UTC](https://forums.developer.nvidia.com/t/nvcr-io-is-down/352247 "2025-11-22T03:16:24Z")
nvcr.io is down. Can you please fix this? --- ## [NVIDIA Driver was not detected inside docker container](https://forums.developer.nvidia.com/t/nvidia-driver-was-not-detected-inside-docker-container/351291)
**Author:** [@PavanMV](https://forums.developer.nvidia.com/u/PavanMV)\ **Replies:** 1\ **Last updated:** [November 20, 2025, 9:55am UTC](https://forums.developer.nvidia.com/t/nvidia-driver-was-not-detected-inside-docker-container/351291 "2025-11-20T09:55:32Z")
Setup: Ubuntu 24/22 docker: Docker version 29.0.0, build 3d4129b nvidia-toolkit-container: 1.18.0/1.17.1 Have already tried : troubleshoot command: docker run --runtime nvidia --gpus all -it nvidia/cuda:11.7.1-cudnn8… --- ## [Audio2face 2d](https://forums.developer.nvidia.com/t/audio2face-2d/351186)
**Author:** [@nikita17](https://forums.developer.nvidia.com/u/nikita17)\ **Replies:** 0\ **Last updated:** [November 12, 2025, 2:10pm UTC](https://forums.developer.nvidia.com/t/audio2face-2d/351186 "2025-11-12T14:10:33Z")
I am trying to activate the subscription on ngc https://org.ngc.nvidia.com/activate getting the below issue\*\* Verification Issue\*\*The serial number(s) you are trying to activate are on hold because you tried them previo… --- ## [No communication between docker and the host](https://forums.developer.nvidia.com/t/no-communication-between-docker-and-the-host/351156)
**Author:** [@ar.gounot](https://forums.developer.nvidia.com/u/ar.gounot)\ **Replies:** 0\ **Last updated:** [November 12, 2025, 9:32am UTC](https://forums.developer.nvidia.com/t/no-communication-between-docker-and-the-host/351156 "2025-11-12T09:32:07Z")
Hello, I just created a container using Isaac ROS, but I’m unable to see the active topics (camera images obtained with Isaac Sim). I followed the tutorial to use fast DDS. The container is launched using net=host and wi… --- ## [NVIDIA L4T JetPack docker container \>36.4.0 release](https://forums.developer.nvidia.com/t/nvidia-l4t-jetpack-docker-container-36-4-0-release/350537)
**Author:** [@KRXK](https://forums.developer.nvidia.com/u/KRXK)\ **Replies:** 0\ **Last updated:** [November 7, 2025, 7:56am UTC](https://forums.developer.nvidia.com/t/nvidia-l4t-jetpack-docker-container-36-4-0-release/350537 "2025-11-07T07:56:28Z")
Hi, We use the L4T jetpack container provided by NVIDIA ( NVIDIA L4T JetPack | NVIDIA NGC ) as the basis for our system images before applying binaries etc. I notice that there has not been a tag since 36.4.0 and wonde… --- ## [Image Pullback Error](https://forums.developer.nvidia.com/t/image-pullback-error/349312)
**Author:** [@alex.callinan](https://forums.developer.nvidia.com/u/alex.callinan)\ **Replies:** 0\ **Last updated:** [October 28, 2025, 2:00pm UTC](https://forums.developer.nvidia.com/t/image-pullback-error/349312 "2025-10-28T14:00:45Z")
I had been using nvcr.io/nvidia/driver:570.148.08-azurelinux3.0 in my AKS cluster as my GPU driver. I am now receiving a pullback error and I can no longer find any trace of the container in documentation. Appreciate … --- ## [Docker service create failure](https://forums.developer.nvidia.com/t/docker-service-create-failure/348366)
**Author:** [@735038964](https://forums.developer.nvidia.com/u/735038964)\ **Replies:** 0\ **Last updated:** [October 20, 2025, 8:47am UTC](https://forums.developer.nvidia.com/t/docker-service-create-failure/348366 "2025-10-20T08:47:06Z")
I have two ORIN modules with custom board. I used sudo docker pull nvcr.io/nvidia/l4t-base:35.4.1 to download the container image. run the docker container successfully： sudo docker run --rm -it --net=host --runtime nv… --- ## [Understanding of how OpenGL libraries are linked to an application within a container using the GPU Operator](https://forums.developer.nvidia.com/t/understanding-of-how-opengl-libraries-are-linked-to-an-application-within-a-container-using-the-gpu-operator/306320)
**Author:** [@sebastien.bonnet.eu](https://forums.developer.nvidia.com/u/sebastien.bonnet.eu)\ **Replies:** 3\ **Last updated:** [September 2, 2025, 7:58am UTC](https://forums.developer.nvidia.com/t/understanding-of-how-opengl-libraries-are-linked-to-an-application-within-a-container-using-the-gpu-operator/306320 "2025-09-02T07:58:36Z")
Hello, I use the GPU operator to deploy OpenGL applications in a Kubernetes Cluster. It works fine, my pod (a basic glxgears) starts and when running nvidia-smi from the nvidia-driver pod I can see that glxgears is usin… --- ## [Does the docker use it's own Filesystem?](https://forums.developer.nvidia.com/t/does-the-docker-use-its-own-filesystem/342167)
**Author:** [@adithya.jvsc](https://forums.developer.nvidia.com/u/adithya.jvsc)\ **Replies:** 0\ **Last updated:** [August 14, 2025, 5:25pm UTC](https://forums.developer.nvidia.com/t/does-the-docker-use-its-own-filesystem/342167 "2025-08-14T17:25:14Z")
I have run python -m site on both host terminal and in the docker running terminals , some of the paths looked the same, so wondered if they use the same python. However i tried checking the list of files in the paths… --- ## [\[Isaac ROS Humble\] apt install fails with dpkg error for libnvidia-compute-560](https://forums.developer.nvidia.com/t/isaac-ros-humble-apt-install-fails-with-dpkg-error-for-libnvidia-compute-560/339850)
**Author:** [@benny6551102](https://forums.developer.nvidia.com/u/benny6551102)\ **Replies:** 2\ **Last updated:** [August 13, 2025, 11:07pm UTC](https://forums.developer.nvidia.com/t/isaac-ros-humble-apt-install-fails-with-dpkg-error-for-libnvidia-compute-560/339850 "2025-08-13T23:07:06Z")
I am encountering a persistent issue when trying to install Isaac ROS DNN packages inside the official Docker container. The installation fails with a dpkg error related to NVIDIA library dependencies, I would like to ad… --- ## [Libnvidia-encode.so is present on host but not available in Docker container with --gpus all](https://forums.developer.nvidia.com/t/libnvidia-encode-so-is-present-on-host-but-not-available-in-docker-container-with-gpus-all/340651)
**Author:** [@ival86](https://forums.developer.nvidia.com/u/ival86)\ **Replies:** 0\ **Last updated:** [July 29, 2025, 8:11pm UTC](https://forums.developer.nvidia.com/t/libnvidia-encode-so-is-present-on-host-but-not-available-in-docker-container-with-gpus-all/340651 "2025-07-29T20:11:03Z")
Hello, I have a system with NVIDIA driver 570.x installed on Ubuntu 20.04. The file /usr/lib/x86\_64-linux-gnu/libnvidia-encode.so is present on the host and nvidia-smi works both on the host and inside containers. I u… --- ## [Failed to fetch https:.../Packages.gz File has unexpected size (1804439 != 1800926). Mirror sync in progress?](https://forums.developer.nvidia.com/t/failed-to-fetch-https-packages-gz-file-has-unexpected-size-1804439-1800926-mirror-sync-in-progress/338600)
**Author:** [@naumanhsa965](https://forums.developer.nvidia.com/u/naumanhsa965)\ **Replies:** 2\ **Last updated:** [July 25, 2025, 7:23am UTC](https://forums.developer.nvidia.com/t/failed-to-fetch-https-packages-gz-file-has-unexpected-size-1804439-1800926-mirror-sync-in-progress/338600 "2025-07-25T07:23:02Z")
Using Docker image ARG CUDA\_IMAGE="12.4.1-devel-ubuntu22.04" FROM nvidia/cuda:${CUDA\_IMAGE} RUN apt-get update RUN apt-get install -y python3 python3-pip wget git build-essential ... Log Errors: =\> ERROR \[fastapi-vll… --- ## [CUDA compute Packages.gz for Ubuntu 22.04 File has unexpected size](https://forums.developer.nvidia.com/t/cuda-compute-packages-gz-for-ubuntu-22-04-file-has-unexpected-size/340186)
**Author:** [@barrettarcher](https://forums.developer.nvidia.com/u/barrettarcher)\ **Replies:** 1\ **Last updated:** [July 25, 2025, 7:14am UTC](https://forums.developer.nvidia.com/t/cuda-compute-packages-gz-for-ubuntu-22-04-file-has-unexpected-size/340186 "2025-07-25T07:14:31Z")
Hello. I’m trying to build a Docker container and I need to download this file. However, I get a hash and size unexpected error. #5 5.550 E: Failed to fetch https://developer.download.nvidia.com/compute/cuda/repos/ubunt… --- ## [Nvidia docker invalid cross-device link](https://forums.developer.nvidia.com/t/nvidia-docker-invalid-cross-device-link/339866)
**Author:** [@benny6551102](https://forums.developer.nvidia.com/u/benny6551102)\ **Replies:** 0\ **Last updated:** [July 22, 2025, 11:37am UTC](https://forums.developer.nvidia.com/t/nvidia-docker-invalid-cross-device-link/339866 "2025-07-22T11:37:46Z")
I am encountering a persistent issue when trying to install Isaac ROS DNN packages inside the official Docker container. The installation fails with a dpkg error related to NVIDIA library dependencies, I would like to ad… --- ## [Invalid public key for CUDA apt repository](https://forums.developer.nvidia.com/t/invalid-public-key-for-cuda-apt-repository/212901)
**Author:** [@user141301](https://forums.developer.nvidia.com/u/user141301)\ **Replies:** 34\ **Last updated:** [July 18, 2025, 10:49pm UTC](https://forums.developer.nvidia.com/t/invalid-public-key-for-cuda-apt-repository/212901 "2025-07-18T22:49:58Z")
Using the cuda 11.2.2 docker container, apt update’s are no longer successful due to unavailable public keys for the cuda apt repository. See example below. Attempting to load the key manually via: wget -qO- https://de… --- ## [Walkthrough for validated docker pull of MolMIM NIM](https://forums.developer.nvidia.com/t/walkthrough-for-validated-docker-pull-of-molmim-nim/339498)
**Author:** [@mertz](https://forums.developer.nvidia.com/u/mertz)\ **Replies:** 0\ **Last updated:** [July 18, 2025, 4:00pm UTC](https://forums.developer.nvidia.com/t/walkthrough-for-validated-docker-pull-of-molmim-nim/339498 "2025-07-18T16:00:38Z")
I have been attempting to pull the MolMIM NIM down to an AWS EC2 instance (L40S GPU) unsuccessfully so far. Here are the steps I have followed: Downloaded all the prerequisites and set them up according to MolMIM Quick… --- ## [Applications not using GPU inside docker container](https://forums.developer.nvidia.com/t/applications-not-using-gpu-inside-docker-container/285911)
**Author:** [@mooshie00](https://forums.developer.nvidia.com/u/mooshie00)\ **Replies:** 2\ **Last updated:** [July 18, 2025, 7:40am UTC](https://forums.developer.nvidia.com/t/applications-not-using-gpu-inside-docker-container/285911 "2025-07-18T07:40:48Z")
Hi everyone, I’ve been trying to make my GPU be utilized when using any graphical application from inside of the docker container, but with no success so far. Posting this as an issue since I’ve followed the instruction… --- ## [Isaac ROS Dev Docker container missing package?](https://forums.developer.nvidia.com/t/isaac-ros-dev-docker-container-missing-package/338441)
**Author:** [@fahyen](https://forums.developer.nvidia.com/u/fahyen)\ **Replies:** 0\ **Last updated:** [July 8, 2025, 9:05pm UTC](https://forums.developer.nvidia.com/t/isaac-ros-dev-docker-container-missing-package/338441 "2025-07-08T21:05:34Z")
I installed the SDK into my docker image: # ZED SDK COPY src/isaac\_ros\_common/docker/scripts/docker\_install-zed-aarch64.sh opt/zed/install-zed-aarch64.sh RUN chmod +x opt/zed/install-zed-aarch64.sh && \\ opt/zed/… [Next page](https://forums.developer.nvidia.com/c/accelerated-computing/nvidia-gpu-cloud-ngc-users/docker-and-nvidia-docker/33.md?page=1)