MosaicML improves efficiency of neural network training with algorithmic methods that deliver speed, boost quality and reduce cost.
The MosaicML Composer is an open source deep learning library purpose-built to make it easy to add algorithmic methods and compose them together into novel recipes that speed up model training and improve model quality.
- Naveen Rao
- Hanlin Tang
- Michael Carbin
- Discussing algorithmic approaches towards more efficient model training with @Wellesley College #CS #DataScience st… https://t.co/QHF2xLdUOu
- Did you know that we're characterizing our compute efficient methods on GPT workloads on Mosaic Explorer? See here: https://t.co/slTHUXnlzp
- Back at the root of our scientific foundation, where @jefrankle and Prof @mcarbin created the research that inspire… https://t.co/hVawU2ll4V