Skip to content

Fix issue #11: Add comprehensive multi-GPU support#182

Open
LifeJiggy wants to merge 1 commit intoQwenLM:mainfrom
LifeJiggy:fix-issue-11-multi-gpu
Open

Fix issue #11: Add comprehensive multi-GPU support#182
LifeJiggy wants to merge 1 commit intoQwenLM:mainfrom
LifeJiggy:fix-issue-11-multi-gpu

Conversation

@LifeJiggy
Copy link

@LifeJiggy LifeJiggy commented Oct 14, 2025

  • Add multi_gpu_support.py with device_map and DataParallel support
  • Implement automatic GPU distribution for component placement
  • Add gradient checkpointing and xFormers memory efficient attention
  • Include automatic mixed precision (AMP) for performance
  • Provide readiness checks and usage examples

Addresses the question 'Is there a plan to support multiple GPUs?' with
full implementation supporting dual GPUs and beyond for faster inference."

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant