Skip to content
This repository was archived by the owner on Jul 30, 2025. It is now read-only.

Conversation

@TimPietrusky
Copy link

Motivation

  • Updated base image to use CUDA 11.8 for better compatibility with modern GPUs
  • Updated Python dependencies to work with Python 3.11
  • Locked dependency versions to ensure build stability:
    • numpy==1.24.3
    • torch==2.0.1+cu118
    • torchvision==0.15.2+cu118
    • runpod~=1.7.13
  • Improved Docker build process with proper dependency installation order

Changes

  • Updated Dockerfile to use runpod/base:0.4.2-cuda11.8.0
  • Updated requirements.txt with compatible package versions
  • Ensured proper CUDA support through PyTorch packages
  • Updated documentation to reflect the changes

- Update base image to runpod/base:0.4.2-cuda11.8.0
- Update Python dependencies to be compatible with Python 3.11
- Lock numpy version to ensure compatibility
- Update torch and torchvision to CUDA 11.8 compatible versions
- Update runpod SDK to latest stable version
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants