Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
fix: AccumulateGrad stream mismatch warning when using DDP with Fabric & Trainer (#21746)
* update * update * update * update * update * update * update * update * Apply suggestion from @deependujha
D
Deependu committed
35e56ef93582c60c8ec5ca2bf1025e7c414d6bb6
Parent: 2849907
Committed by GitHub <noreply@github.com>
on 6/1/2026, 1:12:43 PM