Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
COMMITS
/ examples/pytorch/tensor_parallel/parallelism.py May 20, 2024
A
Enable loss-parallel in example (#19882)
awaelchli committed
May 19, 2024
A
(6/n) Support 2D Parallelism - Trainer example (#19879)
awaelchli committed