Intelligence

Compute efficiency by workload · MLPerf benchmarks × market pricing · 2026-05-21

LLM Inference
Generative text inference — tokens produced per second per GPU.
8,103,115
tok/$ · best on-demand
RunPod · NVIDIA B200-SXM-180GB
NLP
Extractive language understanding — queries answered per second per GPU.
10,967,960
samp/$ · best on-demand
RunPod · NVIDIA H100-SXM-80GB
Vision
Image classification throughput — images classified per second per GPU.
107,253,512
samp/$ · best on-demand
RunPod · NVIDIA H100-SXM-80GB
Recommendation
Recommendation inference — user-item scoring queries per second per GPU.
90,618,261
samp/$ · best on-demand
RunPod · NVIDIA H100-SXM-80GB
Medical Imaging
3D medical image segmentation — volumes processed per second per GPU.
7,871
samp/$ · best on-demand
RunPod · NVIDIA H100-SXM-80GB