Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Enable windows in unittest `test_fetching_is_profiled` (#21479)
* fix unittest `test_fetching_is_profiled`. change simple profiler using time.monotonic to time.perf_counter * fix test --------- Co-authored-by: Bhimraj Yadav <[email protected]>
G
GdoongMathew committed
be2e8d01dd85a03ba30c409c847ac35d3b029502
Parent: f2da71f
Committed by GitHub <[email protected]>
on 1/12/2026, 10:01:59 AM