> browse/registry
$ loading registry▋
found 2 skills in registry
> 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.
#Megatron-Core#Large-Scale Training#NVIDIA
Orchestra-Research> deepspeed
Expert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, sparse attention
#DeepSpeed#Distributed Training#ZeRO
Orchestra-Research