training-llms-megatron

Q: How to install training-llms-megatron?

Open your MCP config: ~/.claude.json. Then add the config to "mcpServers": { "training-llms-megatron": { "command": "...", "args": [] } }. Finally, /mcp in Claude Code.

Q: What is training-llms-megatron best for?

training-llms-megatron is categorized under General. It covers: ai-&-ml, research.

New

SmitheryGeneralby davila7

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.

First seen 5/22/2026

Install & Usage

Open your MCP config

~/.claude.json

Add the server config

Add the configuration to "mcpServers": { "training-llms-megatron": { "command": "...", "args": [] } }

Restart Claude Code

/mcp

View source on GitHub

ai-&-mlresearch

Security Audits

LicenseUnknownSourceWarnRepositoryPass

Frequently Asked Questions

What is training-llms-megatron?

How to install training-llms-megatron?

To install training-llms-megatron: open your mcp config (~/.claude.json), then add the config to "mcpServers": { "training-llms-megatron": { "command": "...", "args": [] } }. Finally, /mcp in Claude Code.

What is training-llms-megatron best for?

training-llms-megatron is a mcp categorized under General. It is designed for: ai-&-ml, research. Created by davila7.