Skip to content

added public mlperf inference v5.0 results in the CM/CMX format #29

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ctuning-admin merged 2 commits into from
Apr 9, 2025
Merged

added public mlperf inference v5.0 results in the CM/CMX format #29

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
8322b3e
Merge pull request #1 from mlcommons/main
gfursin Apr 9, 2025
e279f1a
added public mlperf inference v5.0 results
Apr 9, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
274 changes: 274 additions & 0 deletions ...erence--all--datacenter--closed--graph--offline/2025-04-09.13-33-59.057287/cm-result.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,274 @@
[
{
"Accuracy": "acc: 72.869",
"Availability": "available",
"Division": "closed",
"Location": "closed/NVIDIA/results/H200-SXM-141GBx8_TRT/rgat/Offline",
"MlperfModel": "rgat",
"Model": "rgat",
"Organization": "NVIDIA",
"Platform": "H200-SXM-141GBx8_TRT",
"Result": 441817,
"Result_Units": "Samples/s",
"Scenario": "Offline",
"SystemName": "NVIDIA H200 (8x H200-SXM-141GB, TensorRT)",
"SystemType": "datacenter",
"Units": "Samples/s",
"accelerator_model_name": "NVIDIA H200-SXM-141GB",
"accelerators_per_node": 8,
"compliance": 0,
"errors": 0,
"framework": "TensorRT 10.8, CUDA 12.8",
"git_url": "https://github.com/mlcommons/inference_results_v5.0",
"has_power": false,
"host_processor_core_count": 56,
"host_processor_model_name": "Intel(R) Xeon(R) Platinum 8480C",
"host_processors_per_node": 2,
"inferred": 0,
"notes": "H200 TGP 700W",
"number_of_nodes": 1,
"operating_system": "Ubuntu 24.04",
"uid": "adcd6ba5a59c466e",
"url": "https://github.com/mlcommons/inference_results_v5.0/tree/master/closed/NVIDIA/results/H200-SXM-141GBx8_TRT/rgat/Offline",
"version": "v5.0",
"weight_data_types": "fp16"
},
{
"Accuracy": "acc: 72.860",
"Availability": "available",
"Division": "closed",
"Location": "closed/Oracle/results/H200-SXM-141GBx8_TRT/rgat/Offline",
"MlperfModel": "rgat",
"Model": "rgat",
"Organization": "Oracle",
"Platform": "H200-SXM-141GBx8_TRT",
"Result": 410017,
"Result_Units": "Samples/s",
"Scenario": "Offline",
"SystemName": "NVIDIA H200 (8x H200-SXM-141GB, TensorRT)",
"SystemType": "datacenter",
"Units": "Samples/s",
"accelerator_model_name": "NVIDIA H200-SXM-141GB",
"accelerators_per_node": 8,
"compliance": 0,
"errors": 0,
"framework": "TensorRT 10.8, CUDA 12.8",
"git_url": "https://github.com/mlcommons/inference_results_v5.0",
"has_power": false,
"host_processor_core_count": 56,
"host_processor_model_name": "Intel(R) Xeon(R) Platinum 8480C",
"host_processors_per_node": 2,
"inferred": 0,
"notes": "H200 TGP 700W",
"number_of_nodes": 1,
"operating_system": "Ubuntu 22.04.5 LT",
"uid": "d2615660e21c4ef8",
"url": "https://github.com/mlcommons/inference_results_v5.0/tree/master/closed/Oracle/results/H200-SXM-141GBx8_TRT/rgat/Offline",
"version": "v5.0",
"weight_data_types": "fp16"
},
{
"Accuracy": "acc: 72.853",
"Availability": "available",
"Division": "closed",
"Location": "closed/GigaComputing/results/G893-SD1_H200-SXM-141GBx8_TRT/rgat/Offline",
"MlperfModel": "rgat",
"Model": "rgat",
"Organization": "GigaComputing",
"Platform": "G893-SD1_H200-SXM-141GBx8_TRT",
"Result": 418270,
"Result_Units": "Samples/s",
"Scenario": "Offline",
"SystemName": "G893-SD1",
"SystemType": "datacenter",
"Units": "Samples/s",
"accelerator_model_name": "NVIDIA H200-SXM-141GB",
"accelerators_per_node": 8,
"compliance": 0,
"errors": 0,
"framework": "TensorRT 10.8, CUDA 12.8",
"git_url": "https://github.com/mlcommons/inference_results_v5.0",
"has_power": false,
"host_processor_core_count": 56,
"host_processor_model_name": "Intel(R) Xeon(R) Platinum 8480+",
"host_processors_per_node": 2,
"inferred": 0,
"notes": "H200 TGP 700W",
"number_of_nodes": 1,
"operating_system": "Ubuntu 22.04.4",
"uid": "31650d587de642dc",
"url": "https://github.com/mlcommons/inference_results_v5.0/tree/master/closed/GigaComputing/results/G893-SD1_H200-SXM-141GBx8_TRT/rgat/Offline",
"version": "v5.0",
"weight_data_types": "fp16"
},
{
"Accuracy": "acc: 72.533",
"Availability": "available",
"Division": "closed",
"Location": "closed/Intel/results/1-node-2S-GNR_128C/rgat/Offline",
"MlperfModel": "rgat",
"Model": "rgat",
"Organization": "Intel",
"Platform": "1-node-2S-GNR_128C",
"Result": 24598.2,
"Result_Units": "Samples/s",
"Scenario": "Offline",
"SystemName": "1-node-2S-GNR_128C",
"SystemType": "datacenter",
"Units": "Samples/s",
"accelerator_model_name": "N/A",
"accelerators_per_node": 0,
"compliance": 0,
"errors": 0,
"framework": "PyTorch",
"git_url": "https://github.com/mlcommons/inference_results_v5.0",
"has_power": false,
"host_processor_core_count": 128,
"host_processor_model_name": "INTEL(R) XEON(R) 6980P",
"host_processors_per_node": 2,
"inferred": 0,
"notes": "N/A. N/A",
"number_of_nodes": 1,
"operating_system": "CentOS Stream 9",
"uid": "3dd27c0e231b47c2",
"url": "https://github.com/mlcommons/inference_results_v5.0/tree/master/closed/Intel/results/1-node-2S-GNR_128C/rgat/Offline",
"version": "v5.0",
"weight_data_types": "INT8"
},
{
"Accuracy": "acc: 72.881",
"Availability": "available",
"Division": "closed",
"Location": "closed/Supermicro/results/SYS_821GE_TNHR_H200_SXM_141GBX8_TRT/rgat/Offline",
"MlperfModel": "rgat",
"Model": "rgat",
"Organization": "Supermicro",
"Platform": "SYS_821GE_TNHR_H200_SXM_141GBX8_TRT",
"Result": 430566,
"Result_Units": "Samples/s",
"Scenario": "Offline",
"SystemName": "SYS-821GE-TNHR (8x H200-SXM-141GB, TensorRT)",
"SystemType": "datacenter",
"Units": "Samples/s",
"accelerator_model_name": "NVIDIA H200-SXM-141GB",
"accelerators_per_node": 8,
"compliance": 0,
"errors": 0,
"framework": "TensorRT 10.8, CUDA 12.8",
"git_url": "https://github.com/mlcommons/inference_results_v5.0",
"has_power": false,
"host_processor_core_count": 48,
"host_processor_model_name": "Intel(R) Xeon(R) Platinum 8568Y+",
"host_processors_per_node": 2,
"inferred": 0,
"notes": "H200 TGP 700W",
"number_of_nodes": 1,
"operating_system": "Ubuntu 24.04",
"uid": "a3d9d9643e334b00",
"url": "https://github.com/mlcommons/inference_results_v5.0/tree/master/closed/Supermicro/results/SYS_821GE_TNHR_H200_SXM_141GBX8_TRT/rgat/Offline",
"version": "v5.0",
"weight_data_types": "fp32"
},
{
"Accuracy": "acc: 72.861",
"Availability": "available",
"Division": "closed",
"Location": "closed/Supermicro/results/AS_4125GS_TNHR2_LCC_H200_SXM_141GBX8_TRT/rgat/Offline",
"MlperfModel": "rgat",
"Model": "rgat",
"Organization": "Supermicro",
"Platform": "AS_4125GS_TNHR2_LCC_H200_SXM_141GBX8_TRT",
"Result": 440893,
"Result_Units": "Samples/s",
"Scenario": "Offline",
"SystemName": "AS-4125GS-TNHR2-LCC (8x H200-SXM-141GB, TensorRT)",
"SystemType": "datacenter",
"Units": "Samples/s",
"accelerator_model_name": "NVIDIA H200-SXM-141GB",
"accelerators_per_node": 8,
"compliance": 0,
"errors": 0,
"framework": "TensorRT 10.8, CUDA 12.8",
"git_url": "https://github.com/mlcommons/inference_results_v5.0",
"has_power": false,
"host_processor_core_count": 96,
"host_processor_model_name": "AMD EPYC 9654",
"host_processors_per_node": 2,
"inferred": 0,
"notes": "H200 TGP 700W",
"number_of_nodes": 1,
"operating_system": "Ubuntu 24.04",
"uid": "9069a182d8de4d5e",
"url": "https://github.com/mlcommons/inference_results_v5.0/tree/master/closed/Supermicro/results/AS_4125GS_TNHR2_LCC_H200_SXM_141GBX8_TRT/rgat/Offline",
"version": "v5.0",
"weight_data_types": "fp32"
},
{
"Accuracy": "acc: 72.873",
"Availability": "available",
"Division": "closed",
"Location": "closed/Lenovo/results/SR780a_V3_H200SXMx8/rgat/Offline",
"MlperfModel": "rgat",
"Model": "rgat",
"Organization": "Lenovo",
"Platform": "SR780a_V3_H200SXMx8",
"Result": 450175,
"Result_Units": "Samples/s",
"Scenario": "Offline",
"SystemName": "ThinkSystem SR780a V3(8x H200-SXM-141GB, TensorRT)",
"SystemType": "datacenter",
"Units": "Samples/s",
"accelerator_model_name": "NVIDIA H200-SXM-141GB",
"accelerators_per_node": 8,
"compliance": 0,
"errors": 0,
"framework": "TensorRT 10.8, CUDA 12.8",
"git_url": "https://github.com/mlcommons/inference_results_v5.0",
"has_power": false,
"host_processor_core_count": 48,
"host_processor_model_name": "INTEL(R) XEON(R) PLATINUM 8568Y+",
"host_processors_per_node": 2,
"inferred": 0,
"notes": "H200 TGP 700W",
"number_of_nodes": 1,
"operating_system": "Ubuntu 24.04",
"uid": "edf76ef2391b41de",
"url": "https://github.com/mlcommons/inference_results_v5.0/tree/master/closed/Lenovo/results/SR780a_V3_H200SXMx8/rgat/Offline",
"version": "v5.0",
"weight_data_types": "fp32"
},
{
"Accuracy": "acc: 72.855",
"Availability": "available",
"Division": "closed",
"Location": "closed/Lenovo/results/H200-SXM-141GBx8_TRT/rgat/Offline",
"MlperfModel": "rgat",
"Model": "rgat",
"Organization": "Lenovo",
"Platform": "H200-SXM-141GBx8_TRT",
"Result": 450144,
"Result_Units": "Samples/s",
"Scenario": "Offline",
"SystemName": "ThinkSystem SR680a V3(8x H200-SXM-141GB, TensorRT)",
"SystemType": "datacenter",
"Units": "Samples/s",
"accelerator_model_name": "NVIDIA H200-SXM-141GB",
"accelerators_per_node": 8,
"compliance": 0,
"errors": 0,
"framework": "TensorRT 10.8, CUDA 12.8",
"git_url": "https://github.com/mlcommons/inference_results_v5.0",
"has_power": false,
"host_processor_core_count": 48,
"host_processor_model_name": "INTEL(R) XEON(R) PLATINUM 8568Y+",
"host_processors_per_node": 2,
"inferred": 0,
"notes": "H200 TGP 700W",
"number_of_nodes": 1,
"operating_system": "Ubuntu 24.04",
"uid": "35ff0150f7f645bd",
"url": "https://github.com/mlcommons/inference_results_v5.0/tree/master/closed/Lenovo/results/H200-SXM-141GBx8_TRT/rgat/Offline",
"version": "v5.0",
"weight_data_types": "fp32"
}
]
15 changes: 15 additions & 0 deletions experiment/mlperf-inference--all--datacenter--closed--graph--offline/_cm.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
{
"alias": "mlperf-inference--all--datacenter--closed--graph--offline",
"automation_alias": "experiment",
"automation_uid": "a0a2d123ef064bcb",
"tags": [
"mlperf",
"mlperf-inference",
"all",
"datacenter",
"closed",
"graph",
"offline"
],
"uid": "f6aac2c0822647e9"
}
34 changes: 34 additions & 0 deletions ...nter--closed--image-classification--multistream/2023-04-06.00-36-37.103508/cm-result.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,38 @@
[
{
"Accuracy": "acc: 76.064",
"Availability": "available",
"Division": "closed",
"Location": "closed/GATEOverflow/results/RTX4090x2-nvidia-gpu-TensorRT-default_config/resnet50/multistream",
"MlperfModel": "resnet",
"Model": "resnet50",
"Organization": "GATEOverflow",
"Platform": "RTX4090x2-nvidia-gpu-TensorRT-default_config",
"Result": 0.506056,
"Result_Units": "Latency (ms)",
"Scenario": "MultiStream",
"SystemName": "GATE Overflow Intel Sapphire Rapids (2x RTX 4090)",
"SystemType": "datacenter,edge",
"Units": "Latency (ms)",
"accelerator_model_name": "NVIDIA GeForce RTX 4090",
"accelerators_per_node": 2,
"compliance": 0,
"errors": 0,
"framework": "TensorRT",
"git_url": "https://github.com/mlcommons/inference_results_v5.0",
"has_power": false,
"host_processor_core_count": 24,
"host_processor_model_name": "Intel(R) Xeon(R) w7-2495X",
"host_processors_per_node": 1,
"inferred": 0,
"notes": "",
"number_of_nodes": 1,
"operating_system": "Ubuntu 20.04 (linux-6.8.0-52-generic-glibc2.31)",
"uid": "af0b77b5bb984f7b",
"url": "https://github.com/mlcommons/inference_results_v5.0/tree/master/closed/GATEOverflow/results/RTX4090x2-nvidia-gpu-TensorRT-default_config/resnet50/multistream",
"version": "v5.0",
"weight_data_types": "int8"
},
{
"Accuracy": 76.076,
"Accuracy_div_100": 0.76076,
Expand Down
Loading