Cisco · HPF HGX System (32xH100-SXM-80GB, TensorRT)
Llama-2-70B (MLPerf reference) · llama2-70b-99 · Server · run 01a00b49-28ca…
valid · Reported evidence
Measurements
output_token_throughput_tps · reported122,274 tokens_s
Llama-2-70B (MLPerf reference) · llama2-70b-99 · Server · server · 8× NVIDIA H100-SXM-80GB × 4 nodes · mlperf_closed_99_9 · observed 2025-09-04
Identity
modelLlama-2-70B (MLPerf reference)
stackTensorRT 10.8, CUDA 12.8
workloadllama2-70b-99 · Server
topology8× NVIDIA H100-SXM-80GB × 4 nodes
harnessmlperf_loadgen v5.1
quality policymlperf_closed_99_9
comparison keysha256:652af6df2d91bc02…
sha256:11a000693757732cf3cb711…
Launch configuration
dtypefp8
systemHPF HGX System (32xH100-SXM-80GB, TensorRT)
softwareTensorRT 10.8, CUDA 12.8
availabilityavailable
Workload
namellama2-70b-99 · Server
streamingyes
load generationopen_loop
Canonical manifest
Show manifestHide manifest
{
"kind": "ServingRun",
"spec": {
"status": "valid",
"harness": {
"name": "mlperf_loadgen",
"version": "v5.1"
},
"topology": {
"nodeCount": 4,
"acceleratorModel": "NVIDIA H100-SXM-80GB",
"acceleratorVendor": "nvidia",
"acceleratorsPerNode": 8
},
"observedAt": "2025-09-04T19:30:14Z",
"modelDigest": "sha256:8005d6af1339caa02e60709ce2d7a6002ce9c51e436d21bb39dc5b8f93a24b0c",
"measurements": [
{
"unit": "tokens_s",
"value": 122274,
"metric": "output_token_throughput_tps",
"statistic": "reported"
}
],
"sourceNative": {
"source": "mlperf-inference",
"metrics": {
"performance_units": "Tokens/s",
"performance_result": 122274
},
"category": "datacenter",
"division": "closed",
"scenario": "server",
"benchmark": "llama2-70b-99.9",
"externalId": "v5.1/5.1-0016"
},
"qualityPolicy": "mlperf_closed_99_9",
"workloadDigest": "sha256:ee188c862b2e3bbbca29b194db4aa45d0f33b7da2bdc410aa953d53e6ad6a4ac",
"configurationDigest": "sha256:9335e6a1ab65bbc3a17837505d8091248c6d37726e21320ff66284f962bb54e1"
},
"metadata": {
"name": "mlperf-v5-1-5-1-0016",
"title": "Cisco · llama2-70b-99.9 · Server",
"sourceRefs": [
{
"url": "https://github.com/mlcommons/inference_results_v5.1/tree/main/closed/Cisco/results/HGX-H100_H100-SXM-80GBx32_TRT"
}
]
},
"apiVersion": "kernelindex.dev/v1alpha1"
}Cite this result (permalink, digest, access date)
Report an issue with this serving run
MLPerf Inference · closed/datacenter · entry v5.1/5.1-0016 · shown unmodified as publishedAll serving results →JSON