gpu-containers
/gpu-containers
Use when the user needs to configure Docker/container environments for GPU workloads, mentions CUDA, ROCm, GPU passthrough, or asks about running PyTorch/TensorFlow in containers. Configures NVIDIA/AMD GPU runtime access, selects compatible base images, and validates single/multi-GPU container execution.
- Author
- uw-ssec ↗
- Repository
- rse-plugins ↗
- Licence
- BSD-3-Clause
- Last repository change
- 14.10.2026
Assessment and security
- Files in this package
- 6
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 14.10.2026, without references to retired models or APIs.
BSD-3-Clause, otwarta.
What it does and when to use it
Use when the user needs to configure Docker/container environments for GPU workloads, mentions CUDA, ROCm, GPU passthrough, or asks about running PyTorch/TensorFlow in containers. Configures NVIDIA/AMD GPU runtime access, selects compatible base images, and validates single/multi-GPU container execution.
Original author description
Use when the user needs to configure Docker/container environments for GPU workloads, mentions CUDA, ROCm, GPU passthrough, or asks about running PyTorch/TensorFlow in containers. Configures NVIDIA/AMD GPU runtime access, selects compatible base images, and validates single/multi-GPU container execution.
How it works
- Primary Actions
- Quick Checks
- Compatibility Snapshot
- End-to-End Workflow With Validation Gates
- Quick Compose Example (NVIDIA)
- Templates
- Deep References
SKILL.md preview
---
name: gpu-containers
description: Use when the user needs to configure Docker/container environments for GPU workloads, mentions CUDA, ROCm, GPU passthrough, or asks about running PyTorch/TensorFlow in containers. Configures NVIDIA/AMD GPU runtime access, selects compatible base images, and validates single/multi-GPU container execution.
metadata:
references:
- references/nvidia-cuda-setup.md
- references/amd-rocm-setup.md
- references/gpu-validation-and-operations.md
assets:
- assets/cuda-dockerfile.dockerfile
- assets/rocm-dockerfile.dockerfile
---
# GPU Containers
## Primary Actions
- Configure NVIDIA CUDA or AMD ROCm container runtime access.
- Select compatible CUDA/ROCm base images for framework workloads.
- Configure single- and multi-GPU container execution.
- Configure Docker Compose GPU services.
- Troubleshoot driver/runtime/framework compatibility failures.
## Quick Checks
```bash
# Host visibility
nvidia-smi
# NVIDIA container GPU check
docker run --rm --gpus all nvidia/cuda:12.4.1-base-ubuntu22.04 nvidia-smi
# AMD ROCm container check
docker run --rm --device=/dev/kfd --device=/dev/dri rocm/pytorch:latest rocm-smi
```
## Compatibility Snapshot
| CUDA Version | Minimum Driver |
|-------------|----------------|
| 12.4 | 550.54+ |
| 12.1 | 530.30+ |
| 11.8 | 520.61+ |
## End-to-End Workflow With Validation Gates
1. Check host GPU driver/tooling.
2. Install/configure container runtime toolkit.
3. Validate GPU visibility in a base container.
4. Build app image and validate framework GPU access.
5. For multi-GPU jobs, validate allocation and IPC/shared memory settings.
6. If any gate fails, fix compatibility/runtime config and retry from step 1.
## Quick Compose Example (NVIDIA)
```yaml
services:
training:
build: .
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]
```
## Templates
- [assets/cuda-dockerfile.dockerfile](assets/cuda-dockerfile.dockerfile)
- [assets/rocm-dockerfile.dockerfile](assets/rocm-dockerfile.dockerfile)
## Deep References
- NVIDIA CUDA setup and compatibility:
[references/nvidia-cuda-setup.md](references/nvidia-cuda-setup.md)
- AMD ROCm setup and device mapping:
[references/amd-rocm-setup.md](references/amd-rocm-setup.md)
- Validation workflow and failure recovery:
[references/gpu-validation-and-operations.md](references/gpu-validation-and-operations.md)
Files in this package (6)
- SKILL.md 2 KB
- assets/cuda-dockerfile.dockerfile 5 KB
- assets/rocm-dockerfile.dockerfile 5 KB
- references/amd-rocm-setup.md 10 KB
- references/gpu-validation-and-operations.md 1 KB
- references/nvidia-cuda-setup.md 15 KB
Versions
The database has no other skills named gpu-containers.
Database snapshot: 13.11.2026 · package version accd08499c6d (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 91docker-ros2-development
Best practices for Docker-based ROS2 development including multi-stage Dockerfiles, docker-compose for multi-container robotic systems, DDS discovery across containers, GPU passthrough for perception, and dev-vs-deploy container patterns. Use this skill when containerizing ROS2… - 100gpu-server-setup
Prepare a Linux server with NVIDIA or Intel Arc GPUs for neural-network and LLM workloads, or diagnose one that misbehaves. Use when the user asks to set up or prepare a GPU server, install NVIDIA drivers or CUDA on Debian/Ubuntu, bring up Intel Arc (A770, A750, B580, Arc Pro… - 100tao-setup-nvidia-gpu-host
Host setup for TAO GPU backends. Checks and, after user approval, installs minimum-compatible NVIDIA driver, CUDA Toolkit, and NVIDIA Container Toolkit versions for Docker/local-Docker and Kubernetes GPU worker hosts. TAO-wide defaults can be overridden by the selected model's… - 95flox-cuda
CUDA and GPU development with Flox. Use for NVIDIA CUDA setup, GPU computing, deep learning frameworks, cuDNN, and cross-platform GPU/CPU development. - 100pixi-install-nvidia
Use when the user says "use pixi to install <some nvidia tool>" (or similar) and wants NVIDIA/CUDA/GPU packages installed via Pixi (no sudo/apt), e.g., CUDA toolkit pieces, cuDNN/NCCL, PyTorch CUDA builds, RAPIDS. - 100rocm
AMD ROCm GPU computing stack for HIP kernel development and GPU-accelerated library usage. Use when: writing HIP kernels (.hip files), using rocBLAS/rocFFT/rocRAND/rocSOLVER/rocSPARSE/hipBLAS/hipBLASLt/hipTensor/hipSPARSELt/rocALUTION compute libraries, profiling with…