Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Support NVIDIA's Transformer Engine as a precision plugin (#17597)
C
Carlos Mocholí committed
071f85842e2fbec4566cba6dd6b28da9984e9cbd
Parent: d653e4e
Committed by GitHub <[email protected]>
on 7/19/2023, 4:21:58 PM