AMD · QuantaGrid D74A-7U
Llama-2-70B (MLPerf reference) · llama2-70b-99 · Offline · run 01a00b49-1308…
valid · Reported evidence
Measurements
output_token_throughput_tps · reported34,555 tokens_s
Llama-2-70B (MLPerf reference) · llama2-70b-99 · Offline · offline · 8× AMD Instinct MI325X 256GB HBM3e · mlperf_closed_99_9 · observed 2025-09-04
Identity
modelLlama-2-70B (MLPerf reference)
stackvLLM 0.6.5.dev964+mlperf50, Pytorch 2.7.0a0+git3a58512, ROCm 6.3.1
workloadllama2-70b-99 · Offline
topology8× AMD Instinct MI325X 256GB HBM3e
harnessmlperf_loadgen v5.1
quality policymlperf_closed_99_9
comparison keysha256:054a1b825ffc7a8e…
sha256:c41eac2d51e87233541cf7a…
Launch configuration
dtypefp8
notesMachine 2
systemQuantaGrid D74A-7U
softwarevLLM 0.6.5.dev964+mlperf50, Pytorch 2.7.0a0+git3a58512, ROCm 6.3.1
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": "AMD Instinct MI325X 256GB HBM3e",
"acceleratorVendor": "amd",
"acceleratorsPerNode": 8
},
"observedAt": "2025-09-04T19:30:14Z",
"modelDigest": "sha256:8005d6af1339caa02e60709ce2d7a6002ce9c51e436d21bb39dc5b8f93a24b0c",
"measurements": [
{
"unit": "tokens_s",
"value": 34555,
"metric": "output_token_throughput_tps",
"statistic": "reported"
}
],
"sourceNative": {
"source": "mlperf-inference",
"metrics": {
"performance_units": "Tokens/s",
"performance_result": 34555
},
"category": "datacenter",
"division": "closed",
"scenario": "offline",
"benchmark": "llama2-70b-99.9",
"externalId": "v5.1/5.1-0003"
},
"qualityPolicy": "mlperf_closed_99_9",
"workloadDigest": "sha256:65632b1fcd025d277521ab3869f904644b69e0e5322392b1d05ab8a0be7d84b9",
"configurationDigest": "sha256:f6d50351e549a2fed30d17fcb5c64d207e8180a2b6d4d0b17993ec95e1bbc7c9"
},
"metadata": {
"name": "mlperf-v5-1-5-1-0003",
"title": "AMD · llama2-70b-99.9 · Offline",
"sourceRefs": [
{
"url": "https://github.com/mlcommons/inference_results_v5.1/tree/main/closed/AMD/results/8xMI325X_2xEPYC_9575F_2"
}
]
},
"apiVersion": "kernelindex.dev/v1alpha1"
}Cite this result (permalink, digest, access date)
Report an issue with this serving run
MLPerf Inference · closed/datacenter · entry v5.1/5.1-0003 · shown unmodified as publishedAll serving results →JSON