inference/research
verifiedlow · 0.50updated 2026-10-14

MLPerf Inference v5.0 — Cisco C245M8 2x H100 NVL 94GB, Llama-2-70B, TensorRT

benchmark

Fields

id
sources/bench-mlperf-v5-0-cisco-h100nvl-llama2-70b
URL
https://github.com/mlcommons/inference_results_v5.0/tree/main/closed/Cisco/results/C245M8_H100NVL_94GBx2_TRT
Publisher
MLCommons / Cisco
Kind
benchmark
Accessed
2026-10-14
notes
Raw submission logs. TensorRT 10.8, CUDA 12.8, host 2x EPYC 9224. Llama-2-70B FP8, Offline + Server scenarios, tokens/sec verbatim from mlperf_log_summary.txt. Aggregate across 2x H100-NVL-94GB.

Notes

Raw submission logs. TensorRT 10.8, CUDA 12.8, host 2x EPYC 9224. Llama-2-70B FP8, Offline + Server scenarios, tokens/sec verbatim from mlperf_log_summary.txt. Aggregate across 2x H100-NVL-94GB.

Cross-references

This record references no other records.