Skip to content
KernelIndex
Search⌘K
Sign in

Intel · 1-node-4x-BMG-B70

Llama-3.1-8B (MLPerf reference) · llama3.1-8b · Server · run 01a00b4a-4bf1…
valid · Reported evidence
Measurements
output_token_throughput_tps · reported4,170.13 tokens_s

Llama-3.1-8B (MLPerf reference) · llama3.1-8b · Server · server · 4× Intel(R) Arc Pro(R) B70 · mlperf_closed_99 · observed 2026-03-26

Availability: preview.

Identity
modelLlama-3.1-8B (MLPerf reference)
stackPyTorch vLLM
workloadllama3.1-8b · Server
topology4× Intel(R) Arc Pro(R) B70
harnessmlperf_loadgen v6.0
quality policymlperf_closed_99
comparison keysha256:1d49cd34fe44b4ac…
sha256:8839e04007c94b2eee1f64b…

Launch configuration

dtypeint8
notesN/A. N/A
system1-node-4x-BMG-B70
softwarePyTorch vLLM
availabilitypreview

Workload

namellama3.1-8b · Server
streamingyes
load generationopen_loop

Canonical manifest

Show manifest
{
  "kind": "ServingRun",
  "spec": {
    "status": "valid",
    "caveats": [
      "Availability: preview"
    ],
    "harness": {
      "name": "mlperf_loadgen",
      "version": "v6.0"
    },
    "topology": {
      "nodeCount": 1,
      "acceleratorModel": "Intel(R) Arc Pro(R) B70",
      "acceleratorsPerNode": 4
    },
    "observedAt": "2026-03-26T19:39:54Z",
    "modelDigest": "sha256:dafdb8bb411a12b698d36856eed69f16242fb467999fb68680ca09a5601b6bd4",
    "measurements": [
      {
        "unit": "tokens_s",
        "value": 4170.13,
        "metric": "output_token_throughput_tps",
        "statistic": "reported"
      }
    ],
    "sourceNative": {
      "source": "mlperf-inference",
      "metrics": {
        "performance_units": "Tokens/s",
        "performance_result": 4170.13
      },
      "category": "datacenter",
      "division": "closed",
      "scenario": "server",
      "benchmark": "llama3.1-8b",
      "externalId": "v6.0/6.0-0101"
    },
    "qualityPolicy": "mlperf_closed_99",
    "workloadDigest": "sha256:209ab3ba41ef397171c081377f84bcb2512e3d47cd6334278a712a3858d8eedd",
    "configurationDigest": "sha256:e22ade03534687aa17d8b067b3b0acdbf4097995cea71edbe614e212ac20da99"
  },
  "metadata": {
    "name": "mlperf-v6-0-6-0-0101",
    "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-B70"
      }
    ]
  },
  "apiVersion": "kernelindex.dev/v1alpha1"
}
Cite this result (permalink, digest, access date)
Report an issue with this serving run

A maintainer reviews every report. If accepted, the record is retracted or superseded; its history stays visible.

MLPerf Inference · closed/datacenter · entry v6.0/6.0-0101 · shown unmodified as publishedAll serving results →JSON