NVIDIA Investor Presentation Deck

1x 10x 8x 6x 4x 2x Ox 0.3X Medical Imaging 3D U-Net 0.8X NVIDIA TOPS MLPERF AI INFERENCE BENCHMARKS A100 Is Up To 237X Faster Than the CPU ■Xilinx U250 (Available) NVIDIA T4 (Available) 0.2X OFFLINE Per-Accelerator Normalized to T4 XX Image Speech Classification Recognition ResNet-50 RNN-T Intel Cooper Lake (Preview) NVIDIA A100 (Available) 0.2X X Object Detection SSD-Large X0.1X Recommendation DLRM XX NLP BERT 1x 10x 8x 6x 4x 2x Ox 0.7X 0.2X Image Classification ResNet-50 Per-Accelerator Normalized to T4 ■Xilinx U250 (Available) ■ NVIDIA T4 (Available) SERVER XX Speech Recognition RNN-T X 0.1X Intel Cooper Lake (Preview) ■ NVIDIA A100 (Available) Object Detection SSD-Large 237X vs. CPU X 0.04X Recommendation DLRM XX NLP BERT XX = No result submitted MLPerf v0.7 Inference Closed; Per-accelerator performance derived from the best MLPerf results for respective submissions using reported accelerator count in Data Center Offline and Server. 3D U-Net 99.9%, : 0.7-125, 0.7-113, 0.7-111, ResNet-50: 0.7-119, 0.7-124, 0.7-113, 0.7-111, SSD-Large: 0.7-123, 0.7-113, 0.7-111 DLRM 99.9%: 0.7-126, 0.7-113, 0.7-111, RNN-T, BERT 99.9%: 0.7-111, 0.7-113 MLPerf name and logo are trademarks. See www.mlperf.org for more information. NVIDIA

View entire presentation