feat: add Dockerfile for GPU training with Megatron backend#557
Open
saurabhbikram wants to merge 1 commit intoOpenPipe:mainfrom
Open
feat: add Dockerfile for GPU training with Megatron backend#557saurabhbikram wants to merge 1 commit intoOpenPipe:mainfrom
saurabhbikram wants to merge 1 commit intoOpenPipe:mainfrom
Conversation
Adds a production Dockerfile based on pytorch/pytorch:2.9.0-cuda12.8 with all dependencies for Megatron-based RL training: - megatron-core + megatron-bridge for distributed training - flash-attn compiled from source to match container torch ABI - Transformer Engine rebuilt from source (fixes PyPI wheel ABI mismatch) - grouped_gemm for MoE LoRA support - Workaround for TE triton kernel incompatibility with Triton 3.5+ Also improves .dockerignore to exclude .git/, .github/, .claude/, .ruff_cache/, *.pyc, and example data/venv/wandb directories. https://claude.ai/code/session_017Y9KNNQX2RyVWnqpj3A4hh
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds a production Dockerfile for running Megatron-based RL training on GPU instances. Based on
pytorch/pytorch:2.9.0-cuda12.8-cudnn9-develwith:core.get_int_dtype()not hashable by JIT)Also improves
.dockerignoreto exclude.git/,.github/,.claude/,.ruff_cache/,*.pyc, and example data/venv/wandb directories from the build context.Files changed
Dockerfile(new) — multi-stage GPU training image.dockerignore— additional exclusions for cleaner build context