Skip to content

Conversation

@btaanish
Copy link

This PR introduces the necessary scripts and documentation to enable multi-node execution for the NVIDIA implementation of the Llama2-70B MLPerf Inference benchmark.

The added materials detail configuration, startup instructions, and node-to-node communication setup required for distributed inference across multiple H100 nodes.

✅ PR Checklist – Previous Submission Round repo (v5.0)
This repository contains finalized and published results for a previous submission round. Before submitting changes, please confirm the following:

🔒 Safeguard Against Accidental Disclosure
[ ✅ ] I confirm that I am not committing results or system updates intended for a future round.
[ ✅ ] This PR does not contain any result, metadata, or logs for an unreleased submission round.
🛠️ Valid Post-Publication Changes (if any)
[✅ ] This PR does not alter the published accuracy, latency, or system identity of any prior result.
📄 PR Communication
[✅ ] I have clearly explained the reason for this change in the PR description.

Added instructions for multinode runs and accuracy checks.
@btaanish btaanish requested review from a team as code owners November 10, 2025 14:52
@github-actions
Copy link

github-actions bot commented Nov 10, 2025

MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅

@btaanish
Copy link
Author

recheck

@anandhu-eng anandhu-eng changed the base branch from main to for-scc-2025 November 13, 2025 08:02
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants