Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
ci: bump next `torch` 2.7 (#20656)
* ci: bump next `torch` 2.7 * cuda_version: "12.6.3"
J
Jirka Borovec committed
28f0fcb2861fc3d0e45d6bef858db27a93559726
Parent: 08266a9
Committed by GitHub <noreply@github.com>
on 3/19/2025, 2:15:38 PM