Intel · 1-node-4x-BMG-Pro-B60-Dual
Llama-3.1-8B (MLPerf reference) · llama3.1-8b · Server · run 01a00b4a-1485…
valid · Reported evidence
Measurements
output_token_throughput_tps · reported6,293.89 tokens_s
Llama-3.1-8B (MLPerf reference) · llama3.1-8b · Server · server · 4× MS-Intel Arc Pro B60 Dual 48G Turbo · mlperf_closed_99 · observed 2026-03-26
Identity
modelLlama-3.1-8B (MLPerf reference)
stackPyTorch vLLM
workloadllama3.1-8b · Server
topology4× MS-Intel Arc Pro B60 Dual 48G Turbo
harnessmlperf_loadgen v6.0
quality policymlperf_closed_99
comparison keysha256:4c09c6425c8597ea…
sha256:fe2e3c7e9769cdc835be6da…
Launch configuration
dtypeint8
notesN/A. N/A
system1-node-4x-BMG-Pro-B60-Dual
softwarePyTorch vLLM
availabilityavailable
Workload
namellama3.1-8b · 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": "MS-Intel Arc Pro B60 Dual 48G Turbo",
"acceleratorsPerNode": 4
},
"observedAt": "2026-03-26T19:39:54Z",
"modelDigest": "sha256:dafdb8bb411a12b698d36856eed69f16242fb467999fb68680ca09a5601b6bd4",
"measurements": [
{
"unit": "tokens_s",
"value": 6293.89,
"metric": "output_token_throughput_tps",
"statistic": "reported"
}
],
"sourceNative": {
"source": "mlperf-inference",
"metrics": {
"performance_units": "Tokens/s",
"performance_result": 6293.89
},
"category": "datacenter",
"division": "closed",
"scenario": "server",
"benchmark": "llama3.1-8b",
"externalId": "v6.0/6.0-0060"
},
"qualityPolicy": "mlperf_closed_99",
"workloadDigest": "sha256:209ab3ba41ef397171c081377f84bcb2512e3d47cd6334278a712a3858d8eedd",
"configurationDigest": "sha256:ad75c1618c92919898b6a1e9795c268de92f0314a3729af72af5435780ca7bfc"
},
"metadata": {
"name": "mlperf-v6-0-6-0-0060",
"title": "Intel · llama3.1-8b · Server",
"sourceRefs": [
{
"url": "https://github.com/mlcommons/submissions_inference_v6.0/tree/main/closed/Intel/results/1-node-4x-BMG-Pro-B60-Dual"
}
]
},
"apiVersion": "kernelindex.dev/v1alpha1"
}Cite this result (permalink, digest, access date)
Report an issue with this serving run
MLPerf Inference · closed/datacenter · entry v6.0/6.0-0060 · shown unmodified as publishedAll serving results →JSON