# Megatron-LM **Entity class:** Distributed model framework **As of:** November 17, 2026 ## Definition **Megatron-LM** is NVIDIA's model-parallel training lineage, represented by the Megatron Core parallelism documentation. This node belongs to the **distributed training** route of [[wiki/Distributed Compute Ecosystem|Distributed Compute Ecosystem]], which records the specific kind of work or network capacity involved. ## Relationships - **comparison:** [[wiki/FSDP|FSDP]]. - **mechanism route:** [[wiki/Distributed Compute Ecosystem|Distributed Compute Ecosystem]]. - **source article:** [[articles/Embedded Super-Spreader Events and Muse|Embedded Super-Spreader Events and Muse]]. ## Sources - [Megatron-LM — project record or attributed account](https://docs.nvidia.com/megatron-core/developer-guide/latest/user-guide/parallelism-guide.html).