Google · NVIDIA DGX B200 (8x B200-SXM-180GB, TensorRT)
Llama-3.1-405B (MLPerf reference) · llama3.1-405b · Server · run 01a00b49-eafa…
valid · Reported evidence
Measurements
output_token_throughput_tps · reported983.626 tokens_s
Llama-3.1-405B (MLPerf reference) · llama3.1-405b · Server · server · 8× NVIDIA B200-SXM-180GB · mlperf_closed_99 · observed 2026-03-26
Identity
modelLlama-3.1-405B (MLPerf reference)
stackTensorRT 10.14, CUDA 13.1
workloadllama3.1-405b · Server
topology8× NVIDIA B200-SXM-180GB
harnessmlperf_loadgen v6.0
quality policymlperf_closed_99
comparison keysha256:a9850d9a2ff102bf…
sha256:2132d77495538ec941bbbe3…
Launch configuration
dtypefp4
notesB200 TGP 1000W
systemNVIDIA DGX B200 (8x B200-SXM-180GB, TensorRT)
softwareTensorRT 10.14, CUDA 13.1
availabilityavailable
Workload
namellama3.1-405b · Server
streamingyes
load generationopen_loop
Canonical manifest
Show manifestHide manifest
{
"kind": "ServingRun",
"spec": {
"status": "valid",
"harness": {
"name": "mlperf_loadgen",
"version": "v6.0"
},
"topology": {
"nodeCount": 1,
"acceleratorModel": "NVIDIA B200-SXM-180GB",
"acceleratorVendor": "nvidia",
"acceleratorsPerNode": 8
},
"observedAt": "2026-03-26T19:39:54Z",
"modelDigest": "sha256:bc69ffe83643f1c91a137a177c5ef85829a1f5b65ae51771adb75ef783baf031",
"measurements": [
{
"unit": "tokens_s",
"value": 983.626,
"metric": "output_token_throughput_tps",
"statistic": "reported"
}
],
"sourceNative": {
"source": "mlperf-inference",
"metrics": {
"performance_units": "Tokens/s",
"performance_result": 983.626
},
"category": "datacenter",
"division": "closed",
"scenario": "server",
"benchmark": "llama3.1-405b",
"externalId": "v6.0/6.0-0040"
},
"qualityPolicy": "mlperf_closed_99",
"workloadDigest": "sha256:3428d964054278646d17c2f6064b4e29982b970b7fa607362cf50b7f7dff994c",
"configurationDigest": "sha256:974ba889aa580094dd784b9603578f1c9d77e9384d2b06abb38ddce1ec8cf84f"
},
"metadata": {
"name": "mlperf-v6-0-6-0-0040",
"title": "Google · llama3.1-405b · Server",
"sourceRefs": [
{
"url": "https://github.com/mlcommons/submissions_inference_v6.0/tree/main/closed/Google/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 v6.0/6.0-0040 · shown unmodified as publishedAll serving results →JSON