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

MLPerf Inference v4.1 — Lenovo SR675 V3 8x H100 NVL, Llama-2-70B + GPT-J, TensorRT

benchmark

Fields

id
sources/bench-mlperf-v4-1-lenovo-8xh100nvl-llama2-70b
URL
https://github.com/mlcommons/inference_results_v4.1/tree/main/closed/Lenovo/results/SR675_V3_8xH100_NVL_TRT
Publisher
MLCommons / Lenovo
Kind
benchmark
Accessed
2026-10-14
notes
Raw submission logs. Llama-2-70B + GPT-J Offline/Server, tokens/sec verbatim. Aggregate across 8x H100-NVL-94GB.

Notes

Raw submission logs. Llama-2-70B + GPT-J Offline/Server, tokens/sec verbatim. Aggregate across 8x H100-NVL-94GB.

Cross-references

This record references no other records.