RT @DrJimFan: Most people haven’t had a chance to use H100 yet. Theoretical FLOPs mean nothing - true speedup on Transformers is all that matters.

How does H100 fare against A100 in real battles?

Results on GPT from a third party @MosaicML: the larger model you train, the more time you save. https://t.co/EMR3u4Tjs9