home/categories/machine-learning/orchestra-research-ai-research-skills-08-distributed-training-megatron-core-skill-md
machine-learningdata-ai
training-llms-megatron
Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor/pipeline/sequence/context/expert parallelism. Production-ready framework used for Nemotron, LLaMA, DeepSeek.
maintainer
Orchestra-Research
Updated 11/20/2025
Stars
6563
Forks
515
quick start
Installation and usage
Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor/pipeline/sequence/context/expert parallelism. Production-ready framework used for Nemotron, LLaMA, DeepSeek.
Installation
$ install --globalskills.sh
Usage
Once installed, you can use this skill by running the following command in your terminal:
skills use training-llms-megatron