Home > Workload Solutions > High Performance Computing > White Papers > Dell Validated Design for Government HPC, Artificial Intelligence, and Data Analytics: AI Inferencing Option > Single server performance
The performance metrics for the twenty MLPerf Inference benchmarks on a single server are shown below:
| Server (Queries/s) | Offline (Sample/s) |
resnet50 | 620874 | 702854 |
retinanet | 12484.8 | 14137.9 |
rnnt | 178016 | 184905 |
3D-Unet | 51.9626 | 51.9626 |
bert 99 | 57022.2 | 70129.1 |
bert 99.9 | 51217.9 | 62327 |
gptj 99 | 81.2791 | 105.939 |
gptj 99.9 | 81.2791 | 105.939 |
dlrmv2 99 | 319792 | 340689 |
dlrmv2 99.9 | 319792 | 340689 |
These results can be compared to previous runs by Dell and other organizations on MLCommons. For more information, see MLPerf Inference: Datacenter Benchmark Suite Results.
A previous run with the identical Dell hardware has public submission ID “3.1-0069”