MLPerf Inference v4.1 — Dell R760 2x H100 PCIe 80GB, GPT-J 99%, TensorRT

Fields

id

sources/bench-mlperf-v4-1-dell-r760-h100pcie-gptj

URL

https://github.com/mlcommons/inference_results_v4.1/tree/main/closed/Dell/results/R760_H100_PCIe_80GBx2_TRT

Publisher

MLCommons / Dell

Kind

benchmark

Accessed

2026-10-14

notes

Raw submission logs. TensorRT 10.2.0, CUDA 12.4, host 2x Xeon Platinum 8580. GPT-J 6B FP8, Offline + Server scenarios, tokens/sec (inferred) verbatim from mlperf_log_summary.txt. Aggregate across 2x H100-PCIe-80GB.

Notes

Raw submission logs. TensorRT 10.2.0, CUDA 12.4, host 2x Xeon Platinum 8580. GPT-J 6B FP8, Offline + Server scenarios, tokens/sec (inferred) verbatim from mlperf_log_summary.txt. Aggregate across 2x H100-PCIe-80GB.

Cross-references

This record references no other records.