Skip to content

stanford-crfm/mosaicml-benchmarks

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

benchmarks

Fast reference benchmarks for training ML models with recipes. Designed to be easily forked and modified.

ResNet-50 + ImageNet

drawing

Figure 1: Comparison of MosaicML recipes against other results, all measured on 8x A100s on MosaicML Cloud.

Train the MosaicML ResNet, the fastest ResNet50 implementation that yields a ✨ 7x ✨ faster time-to-train compared to a strong baseline. See our blog for more details and recipes. Our recipes were also demonstrated at MLPerf, a cross industry ML benchmark.

🚀 Get started with the code here.

Large Language Models (LLMs)

Training curves for various LLM sizes.

A simple yet feature complete implementation of GPT, that scales to 70B parameters while maintaining high performance on GPU clusters. Flexible code, written with vanilla PyTorch, that uses PyTorch FSDP and some recent efficiency improvements.

🚀 Get started with the code here.

About

Fast and flexible reference benchmarks

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 99.3%
  • Other 0.7%