Dell · PowerEdge XE9785L (8x MI355X)
Llama-2-70B (MLPerf reference) · llama2-70b-99 · Server · run 01a00b49-c8aa…
valid · Reported evidence
Measurements
output_token_throughput_tps · reported96,734.4 tokens_s
Llama-2-70B (MLPerf reference) · llama2-70b-99 · Server · server · 8× AMD Instinct MI355X 288GB HBM3e · mlperf_closed_99 · observed 2026-03-26
Identity
modelLlama-2-70B (MLPerf reference)
stackvLLM 0.14.0rc3.dev30+gb026cf14e.d20260212.rocm700, Pytorch 2.9.1+git8907517, ROCm 7.0.0.70000-38~22.04
workloadllama2-70b-99 · Server
topology8× AMD Instinct MI355X 288GB HBM3e
harnessmlperf_loadgen v6.0
quality policymlperf_closed_99
comparison keysha256:e12dad2c9f57992c…
sha256:b9066c097fc5d3def5ff334…
Launch configuration
dtypefp4
notesMI355X TGP 1400W
systemPowerEdge XE9785L (8x MI355X)
softwarevLLM 0.14.0rc3.dev30+gb026cf14e.d20260212.rocm700, Pytorch 2.9.1+git8907517, ROCm 7.0.0.70000-38~22.04
availabilityavailable
Workload
namellama2-70b-99 · 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": "AMD Instinct MI355X 288GB HBM3e",
"acceleratorVendor": "amd",
"acceleratorsPerNode": 8
},
"observedAt": "2026-03-26T19:39:54Z",
"modelDigest": "sha256:8005d6af1339caa02e60709ce2d7a6002ce9c51e436d21bb39dc5b8f93a24b0c",
"measurements": [
{
"unit": "tokens_s",
"value": 96734.4,
"metric": "output_token_throughput_tps",
"statistic": "reported"
}
],
"sourceNative": {
"source": "mlperf-inference",
"metrics": {
"performance_units": "Tokens/s",
"performance_result": 96734.4
},
"category": "datacenter",
"division": "closed",
"scenario": "server",
"benchmark": "llama2-70b-99",
"externalId": "v6.0/6.0-0020"
},
"qualityPolicy": "mlperf_closed_99",
"workloadDigest": "sha256:ee188c862b2e3bbbca29b194db4aa45d0f33b7da2bdc410aa953d53e6ad6a4ac",
"configurationDigest": "sha256:e19f832b6b23bf820ab79ecc8e3b42d945979a30a4f5136539367c43e1d306f3"
},
"metadata": {
"name": "mlperf-v6-0-6-0-0020",
"title": "Dell · llama2-70b-99 · Server",
"sourceRefs": [
{
"url": "https://github.com/mlcommons/submissions_inference_v6.0/tree/main/closed/Dell/results/8xMI355X_2xEPYC_9655"
}
]
},
"apiVersion": "kernelindex.dev/v1alpha1"
}Cite this result (permalink, digest, access date)
Report an issue with this serving run
MLPerf Inference · closed/datacenter · entry v6.0/6.0-0020 · shown unmodified as publishedAll serving results →JSON