Graphcore has released the first set of performance benchmarks for its latest AI compute systems – the IPU-M2000 and scale-up IPU-POD64. Across a range of popular models, Graphcore technologies significantly outperformed nVidia’s A100 (DGX-based), both at training and inference. Highlights include: Training EfficientNet-B4:      18x higher throughput ResNeXt 101:          3.7x ...

This story continues at Graphcore benchmarks outperform Nvidia

Or just read more coverage at Electronics Weekly