Mcore Run On Slurm

by nvidiacf5224d14250Apache-2.03.5K starsListed Oct 8, 2026Updated Oct 8, 2026Repository updated today

How to launch distributed Megatron-LM training jobs on a SLURM cluster. Covers a minimal sbatch skeleton, environment-variable setup for torch.distributed.run, CUDA_DEVICE_MAX_CONNECTIONS rules across hardware and parallelism modes, container conventions, monitoring, and per-rank failure diagnosis.

Instructions onlyDevOps & Cloud
  1. cf5224d14250Currentcommit cf5224dPublished Oct 8, 2026

Source and attribution

Source:nvidia/skillsinskills/mcore-run-on-slurmat commitcf5224d

License: Apache-2.0

Content belongs to its original authors. SourceWeft indexes it from a public repository.

Report or request removal