Run python in a cloud-first AI-Dock container. Nothing is added to the installed python environment(s) - You'll have python & pip.
This image provides a great starting point for python development when used standalone but its also a solid foundation for extending upon.
- I built this directly with
docker build --build-arg PYTHON_VERSION=3.12 --build-arg PYTHON_VENV_NAME=python_312 -t mgagliardi/python:v2-py312 . - There's a reliance in this build on the source image having some particular env vars (ex.
VENV_DIR,XPU_TARGET) exposed. Need to pay attention to this if things aren't working. - In
common.shI changedpython3.10topython3.12to correspond with the intent of the image. - I pushed up to Docker Hub with
docker push mgagliardi/python:v2-py312
All AI-Dock containers share a common base which is designed to make running on cloud services such as vast.ai and runpod.io as straightforward and user friendly as possible.
Common features and options are documented in the base wiki but any additional features unique to this image will be detailed below.
The :latest tag points to :latest-cuda
Tags follow these patterns:
-
:[python-version]-v2-cuda-[x.x.x]-[base|runtime|devel]-[ubuntu-version] -
:latest-cuda→:3.10-v2-cuda-11.8.0-cudnn8-runtime-22.04
-
:[python-version]-v2-rocm-[x.x.x]-[core|runtime]-[ubuntu-version] -
:latest-rocm→:3.10-v2-rocm-6.0-runtime-22.04
ROCm builds are experimental. Please give feedback.
-
:[python-version]-v2-cpu-[ubuntu-version] -
:latest-cpu→:3.10-v2-cpu-22.04
Browse here for an image suitable for your target environment.
Supported Python versions: 3.10
Supported Platforms: NVIDIA CUDA, AMD ROCm, CPU
Note
Recent builds include v2 in their image tag. These images use venv rather than micromamba for environment management.
Vast.ai
python:latest-cuda (CUDA)
python:latest-rocm (ROCm)
Runpod.io
Note
These templates are configured to use the latest tag but you are free to change to any of the available Python CUDA tags listed here
The author (@robballantyne) may be compensated if you sign up to services linked in this document. Testing multiple variants of GPU images in many different environments is both costly and time-consuming; This along with sponsorships helps to offset costs and further the development of the project