training-mfu-calculator

Community

Evaluate large model training efficiency with precise MFU metrics.

Authordongg622
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill provides an automated way to compute the Model FLOPs Utilization (MFU) for large-scale neural network training, enabling users to assess hardware utilization during training sessions.

Core Features & Use Cases

  • MFU Calculation: Accurately estimate hardware efficiency based on model configuration, training parameters, and hardware specs.
  • Performance Analysis: Generate detailed reports on FLOPs, GPU utilization, and throughput for model training workflows.
  • Use Case: A developer trains a 70B parameter model on 128 GPUs; this Skill helps quantify the actual compute utilization to identify optimization opportunities.

Quick Start

Describe the model and training setup, then invoke the tool to produce a comprehensive performance report and MFU value.

Dependency Matrix

Required Modules

None required

Components

scriptsreferences

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: training-mfu-calculator
Download link: https://github.com/dongg622/china-ai-chip-skill/archive/main.zip#training-mfu-calculator

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.