NVIDIA · NVIDIA DGX B200 (8x B200-SXM-180GB, TensorRT)
Llama-2-70B (MLPerf reference) · llama2-70b-99 · Offline · run 01a00b49-6f95…
valid · Reported evidence
Measurements
output_token_throughput_tps · reported101,527 tokens_s
Llama-2-70B (MLPerf reference) · llama2-70b-99 · Offline · offline · 8× NVIDIA B200-SXM-180GB · mlperf_closed_99 · observed 2025-09-04
Identity
modelLlama-2-70B (MLPerf reference)
stackTensorRT 10.11, CUDA 12.9
workloadllama2-70b-99 · Offline
topology8× NVIDIA B200-SXM-180GB
harnessmlperf_loadgen v5.1
quality policymlperf_closed_99
comparison keysha256:5c730c54a43e9ae7…
sha256:b1ec4e044f227157d90029c…
Launch configuration
dtypefp4
notesB200 TGP 1000W
systemNVIDIA DGX B200 (8x B200-SXM-180GB, TensorRT)
softwareTensorRT 10.11, CUDA 12.9
availabilityavailable
Workload
namellama2-70b-99 · Offline
streamingno
load generationoffline
Canonical manifest
Show manifestHide manifest
{
"kind": "ServingRun",
"spec": {
"status": "valid",
"harness": {
"name": "mlperf_loadgen",
"version": "v5.1"
},
"topology": {
"nodeCount": 1,
"acceleratorModel": "NVIDIA B200-SXM-180GB",
"acceleratorVendor": "nvidia",
"acceleratorsPerNode": 8
},
"observedAt": "2025-09-04T19:30:14Z",
"modelDigest": "sha256:8005d6af1339caa02e60709ce2d7a6002ce9c51e436d21bb39dc5b8f93a24b0c",
"measurements": [
{
"unit": "tokens_s",
"value": 101527,
"metric": "output_token_throughput_tps",
"statistic": "reported"
}
],
"sourceNative": {
"source": "mlperf-inference",
"metrics": {
"performance_units": "Tokens/s",
"performance_result": 101527
},
"category": "datacenter",
"division": "closed",
"scenario": "offline",
"benchmark": "llama2-70b-99",
"externalId": "v5.1/5.1-0069"
},
"qualityPolicy": "mlperf_closed_99",
"workloadDigest": "sha256:65632b1fcd025d277521ab3869f904644b69e0e5322392b1d05ab8a0be7d84b9",
"configurationDigest": "sha256:d23edc5717f0c077c26a61590e0a5836dd9f9706ca121bb872caf36d060dec61"
},
"metadata": {
"name": "mlperf-v5-1-5-1-0069",
"title": "NVIDIA · llama2-70b-99 · Offline",
"sourceRefs": [
{
"url": "https://github.com/mlcommons/inference_results_v5.1/tree/main/closed/NVIDIA/results/B200-SXM-180GBx8_TRT"
}
]
},
"apiVersion": "kernelindex.dev/v1alpha1"
}Cite this result (permalink, digest, access date)
Report an issue with this serving run
MLPerf Inference · closed/datacenter · entry v5.1/5.1-0069 · shown unmodified as publishedAll serving results →JSON