MLPerf v6.0 Dell 8x NVIDIA H200-NVL-141GB: llama2-70b-99.9 Offline aggregate tok/s
32004.0 output tokens/s (aggregate over 8 accelerators, Offline)
Fields
id
benchmarks/mlperf-v6-0-dell-8xh200nvl-llama2-70b-llama2-70b-99-9-offlineMetric
tps_aggregate
Value
32,004
Unit
output tokens/s (aggregate over 8 accelerators, Offline)
Model
llama2-70b-99.9
Methodology
VENDOR-SUBMITTED MLPerf Inference v6.0 closed division, Datacenter suite, Offline scenario. Dell 8xH200-NVL-141GB: 8x NVIDIA H200-NVL-141GB. Framework TensorRT-LLM. 'Tokens/s' = AGGREGATE output-token throughput across the 8 accelerators (NOT per-request). Value verbatim from summary_results.json. min_query_count/min_duration per MLPerf rules; binding latency constraint applies to Server/Interactive.
Measured by
vendor
Reproducible
yes