version stringclasses 3
values | id stringclasses 223
values | submitter stringclasses 36
values | accelerator stringclasses 35
values | model stringclasses 25
values | scenario stringclasses 3
values | units stringclasses 3
values | total int64 1 288 | result float64 0.02 2.49M | per_accelerator float64 0 96.7k | availability stringclasses 2
values | dtype stringclasses 18
values | system stringclasses 182
values |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
v5.0 | 5.0-0001 | AMD | AMD Instinct MI325X 256GB HBM3E | llama2-70b-99 | Offline | Tokens/s | 8 | 33,928.1 | 4,241.0125 | available | fp8 | QuantaGrid D74A-7U |
v5.0 | 5.0-0001 | AMD | AMD Instinct MI325X 256GB HBM3E | llama2-70b-99 | Server | Tokens/s | 8 | 30,724.5 | 3,840.5625 | available | fp8 | QuantaGrid D74A-7U |
v5.0 | 5.0-0001 | AMD | AMD Instinct MI325X 256GB HBM3E | llama2-70b-99.9 | Offline | Tokens/s | 8 | 33,928.1 | 4,241.0125 | available | fp8 | QuantaGrid D74A-7U |
v5.0 | 5.0-0001 | AMD | AMD Instinct MI325X 256GB HBM3E | llama2-70b-99.9 | Server | Tokens/s | 8 | 30,724.5 | 3,840.5625 | available | fp8 | QuantaGrid D74A-7U |
v5.0 | 5.0-0002 | AMD | AMD Instinct MI325X 256GB HBM3E | stable-diffusion-xl | Offline | Samples/s | 8 | 17.0963 | 2.137038 | available | fp8,fp16,int8 | Quanta S7PA |
v5.0 | 5.0-0002 | AMD | AMD Instinct MI325X 256GB HBM3E | stable-diffusion-xl | Server | Queries/s | 8 | 16.1757 | 2.021963 | available | fp8,fp16,int8 | Quanta S7PA |
v5.0 | 5.0-0003 | ASUSTeK | NVIDIA H200-NVL-141GB | dlrm-v2-99 | Offline | Samples/s | 8 | 326,670 | 40,833.75 | available | int8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB NVLink, TensorRT) |
v5.0 | 5.0-0003 | ASUSTeK | NVIDIA H200-NVL-141GB | dlrm-v2-99 | Server | Queries/s | 8 | 525,118 | 65,639.75 | available | int8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB NVLink, TensorRT) |
v5.0 | 5.0-0003 | ASUSTeK | NVIDIA H200-NVL-141GB | dlrm-v2-99.9 | Offline | Samples/s | 8 | 325,823 | 40,727.875 | available | int8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB NVLink, TensorRT) |
v5.0 | 5.0-0003 | ASUSTeK | NVIDIA H200-NVL-141GB | dlrm-v2-99.9 | Server | Queries/s | 8 | 300,029 | 37,503.625 | available | int8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB NVLink, TensorRT) |
v5.0 | 5.0-0003 | ASUSTeK | NVIDIA H200-NVL-141GB | gptj-99 | Offline | Tokens/s | 8 | 18,113.4 | 2,264.175 | available | fp8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB NVLink, TensorRT) |
v5.0 | 5.0-0003 | ASUSTeK | NVIDIA H200-NVL-141GB | gptj-99 | Server | Tokens/s | 8 | 17,953.9 | 2,244.2375 | available | fp8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB NVLink, TensorRT) |
v5.0 | 5.0-0003 | ASUSTeK | NVIDIA H200-NVL-141GB | gptj-99.9 | Offline | Tokens/s | 8 | 18,113.4 | 2,264.175 | available | fp8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB NVLink, TensorRT) |
v5.0 | 5.0-0003 | ASUSTeK | NVIDIA H200-NVL-141GB | gptj-99.9 | Server | Tokens/s | 8 | 17,953.9 | 2,244.2375 | available | fp8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB NVLink, TensorRT) |
v5.0 | 5.0-0003 | ASUSTeK | NVIDIA H200-NVL-141GB | llama2-70b-99 | Offline | Tokens/s | 8 | 27,765.3 | 3,470.6625 | available | fp8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB NVLink, TensorRT) |
v5.0 | 5.0-0003 | ASUSTeK | NVIDIA H200-NVL-141GB | llama2-70b-99 | Server | Tokens/s | 8 | 27,417.3 | 3,427.1625 | available | fp8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB NVLink, TensorRT) |
v5.0 | 5.0-0003 | ASUSTeK | NVIDIA H200-NVL-141GB | llama2-70b-99.9 | Offline | Tokens/s | 8 | 27,765.3 | 3,470.6625 | available | fp8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB NVLink, TensorRT) |
v5.0 | 5.0-0003 | ASUSTeK | NVIDIA H200-NVL-141GB | llama2-70b-99.9 | Server | Tokens/s | 8 | 27,417.3 | 3,427.1625 | available | fp8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB NVLink, TensorRT) |
v5.0 | 5.0-0004 | ASUSTeK | NVIDIA H200-NVL-141GB | 3d-unet-99 | Offline | Samples/s | 8 | 49.5439 | 6.192988 | available | int8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB, TensorRT) |
v5.0 | 5.0-0004 | ASUSTeK | NVIDIA H200-NVL-141GB | 3d-unet-99.9 | Offline | Samples/s | 8 | 49.5439 | 6.192988 | available | int8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB, TensorRT) |
v5.0 | 5.0-0004 | ASUSTeK | NVIDIA H200-NVL-141GB | resnet | Offline | Samples/s | 8 | 680,738 | 85,092.25 | available | int8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB, TensorRT) |
v5.0 | 5.0-0004 | ASUSTeK | NVIDIA H200-NVL-141GB | resnet | Server | Queries/s | 8 | 540,125 | 67,515.625 | available | int8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB, TensorRT) |
v5.0 | 5.0-0004 | ASUSTeK | NVIDIA H200-NVL-141GB | retinanet | Offline | Samples/s | 8 | 11,263.8 | 1,407.975 | available | int8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB, TensorRT) |
v5.0 | 5.0-0004 | ASUSTeK | NVIDIA H200-NVL-141GB | retinanet | Server | Queries/s | 8 | 11,988.9 | 1,498.6125 | available | int8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB, TensorRT) |
v5.0 | 5.0-0004 | ASUSTeK | NVIDIA H200-NVL-141GB | stable-diffusion-xl | Offline | Samples/s | 8 | 16.0315 | 2.003938 | available | fp8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB, TensorRT) |
v5.0 | 5.0-0004 | ASUSTeK | NVIDIA H200-NVL-141GB | stable-diffusion-xl | Server | Queries/s | 8 | 14.8098 | 1.851225 | available | fp8 | ASUSTeK ESC8000A-E13P (8x H200-NVL-141GB, TensorRT) |
v5.0 | 5.0-0005 | ASUSTeK | AMD Instinct MI325X 256GB HBM3E | llama2-70b-99 | Offline | Tokens/s | 8 | 33,366.5 | 4,170.8125 | available | Dummy | ASUSTeK ESC A8A-E12U (8x MI325x-256GB ) |
v5.0 | 5.0-0005 | ASUSTeK | AMD Instinct MI325X 256GB HBM3E | llama2-70b-99 | Server | Tokens/s | 8 | 30,086.5 | 3,760.8125 | available | Dummy | ASUSTeK ESC A8A-E12U (8x MI325x-256GB ) |
v5.0 | 5.0-0005 | ASUSTeK | AMD Instinct MI325X 256GB HBM3E | llama2-70b-99.9 | Offline | Tokens/s | 8 | 33,366.5 | 4,170.8125 | available | Dummy | ASUSTeK ESC A8A-E12U (8x MI325x-256GB ) |
v5.0 | 5.0-0005 | ASUSTeK | AMD Instinct MI325X 256GB HBM3E | llama2-70b-99.9 | Server | Tokens/s | 8 | 30,086.5 | 3,760.8125 | available | Dummy | ASUSTeK ESC A8A-E12U (8x MI325x-256GB ) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | gptj-99 | Offline | Tokens/s | 8 | 21,171.3 | 2,646.4125 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | gptj-99 | Server | Tokens/s | 8 | 21,533.9 | 2,691.7375 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | gptj-99.9 | Offline | Tokens/s | 8 | 21,171.3 | 2,646.4125 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | gptj-99.9 | Server | Tokens/s | 8 | 21,533.9 | 2,691.7375 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | llama2-70b-99 | Offline | Tokens/s | 8 | 34,807.1 | 4,350.8875 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | llama2-70b-99 | Server | Tokens/s | 8 | 33,037.2 | 4,129.65 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | llama2-70b-99.9 | Offline | Tokens/s | 8 | 34,807.1 | 4,350.8875 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | llama2-70b-99.9 | Server | Tokens/s | 8 | 33,037.2 | 4,129.65 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | llama2-70b-interactive-99 | Offline | Tokens/s | 8 | 34,809.3 | 4,351.1625 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | llama2-70b-interactive-99 | Server | Tokens/s | 8 | 19,418.5 | 2,427.3125 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | llama2-70b-interactive-99.9 | Offline | Tokens/s | 8 | 34,809.3 | 4,351.1625 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | llama2-70b-interactive-99.9 | Server | Tokens/s | 8 | 19,418.5 | 2,427.3125 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | llama3.1-405b | Offline | Tokens/s | 8 | 557.097 | 69.637125 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0006 | ASUSTeK | NVIDIA H200-SXM-141GB | llama3.1-405b | Server | Tokens/s | 8 | 291.183 | 36.397875 | available | fp8 | ASUSTeK ESC N8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0007 | Broadcom_Supermicro | Virtualized NVIDIA H100-SXM-80GB | 3d-unet-99 | Offline | Samples/s | 8 | 51.3784 | 6.4223 | available | int8 | SYS-821GE-TNRT (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0007 | Broadcom_Supermicro | Virtualized NVIDIA H100-SXM-80GB | 3d-unet-99.9 | Offline | Samples/s | 8 | 51.3784 | 6.4223 | available | int8 | SYS-821GE-TNRT (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0007 | Broadcom_Supermicro | Virtualized NVIDIA H100-SXM-80GB | gptj-99 | Offline | Tokens/s | 8 | 19,936.5 | 2,492.0625 | available | fp8 | SYS-821GE-TNRT (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0007 | Broadcom_Supermicro | Virtualized NVIDIA H100-SXM-80GB | gptj-99 | Server | Tokens/s | 8 | 19,318.9 | 2,414.8625 | available | fp8 | SYS-821GE-TNRT (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0007 | Broadcom_Supermicro | Virtualized NVIDIA H100-SXM-80GB | gptj-99.9 | Offline | Tokens/s | 8 | 19,936.5 | 2,492.0625 | available | fp8 | SYS-821GE-TNRT (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0007 | Broadcom_Supermicro | Virtualized NVIDIA H100-SXM-80GB | gptj-99.9 | Server | Tokens/s | 8 | 19,318.9 | 2,414.8625 | available | fp8 | SYS-821GE-TNRT (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0007 | Broadcom_Supermicro | Virtualized NVIDIA H100-SXM-80GB | retinanet | Offline | Samples/s | 8 | 14,313.3 | 1,789.1625 | available | int8 | SYS-821GE-TNRT (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0007 | Broadcom_Supermicro | Virtualized NVIDIA H100-SXM-80GB | retinanet | Server | Queries/s | 8 | 12,869.8 | 1,608.725 | available | int8 | SYS-821GE-TNRT (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0009 | Cisco | NVIDIA H100-NVL-94GB | 3d-unet-99 | Offline | Samples/s | 2 | 10.825 | 5.4125 | available | int8 | Cisco UCS C245 M8 (2x H100NVL-PCIe-94GB) |
v5.0 | 5.0-0009 | Cisco | NVIDIA H100-NVL-94GB | dlrm-v2-99 | Offline | Samples/s | 2 | 103,047 | 51,523.5 | available | int8 | Cisco UCS C245 M8 (2x H100NVL-PCIe-94GB) |
v5.0 | 5.0-0009 | Cisco | NVIDIA H100-NVL-94GB | dlrm-v2-99 | Server | Queries/s | 2 | 96,987.8 | 48,493.9 | available | int8 | Cisco UCS C245 M8 (2x H100NVL-PCIe-94GB) |
v5.0 | 5.0-0009 | Cisco | NVIDIA H100-NVL-94GB | gptj-99 | Offline | Tokens/s | 2 | 3,789.42 | 1,894.71 | available | fp8 | Cisco UCS C245 M8 (2x H100NVL-PCIe-94GB) |
v5.0 | 5.0-0009 | Cisco | NVIDIA H100-NVL-94GB | gptj-99 | Server | Tokens/s | 2 | 3,305.27 | 1,652.635 | available | fp8 | Cisco UCS C245 M8 (2x H100NVL-PCIe-94GB) |
v5.0 | 5.0-0009 | Cisco | NVIDIA H100-NVL-94GB | llama2-70b-99 | Offline | Tokens/s | 2 | 3,741.32 | 1,870.66 | available | fp8 | Cisco UCS C245 M8 (2x H100NVL-PCIe-94GB) |
v5.0 | 5.0-0009 | Cisco | NVIDIA H100-NVL-94GB | llama2-70b-99 | Server | Tokens/s | 2 | 3,337.55 | 1,668.775 | available | fp8 | Cisco UCS C245 M8 (2x H100NVL-PCIe-94GB) |
v5.0 | 5.0-0009 | Cisco | NVIDIA H100-NVL-94GB | resnet | Offline | Samples/s | 2 | 124,595 | 62,297.5 | available | int8 | Cisco UCS C245 M8 (2x H100NVL-PCIe-94GB) |
v5.0 | 5.0-0009 | Cisco | NVIDIA H100-NVL-94GB | resnet | Server | Queries/s | 2 | 109,989 | 54,994.5 | available | int8 | Cisco UCS C245 M8 (2x H100NVL-PCIe-94GB) |
v5.0 | 5.0-0009 | Cisco | NVIDIA H100-NVL-94GB | retinanet | Offline | Samples/s | 2 | 2,677.13 | 1,338.565 | available | int8 | Cisco UCS C245 M8 (2x H100NVL-PCIe-94GB) |
v5.0 | 5.0-0009 | Cisco | NVIDIA H100-NVL-94GB | retinanet | Server | Queries/s | 2 | 2,400.54 | 1,200.27 | available | int8 | Cisco UCS C245 M8 (2x H100NVL-PCIe-94GB) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | gptj-99 | Offline | Tokens/s | 8 | 20,307.7 | 2,538.4625 | available | fp8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | gptj-99 | Server | Tokens/s | 8 | 20,706.7 | 2,588.3375 | available | fp8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | gptj-99.9 | Offline | Tokens/s | 8 | 20,307.7 | 2,538.4625 | available | fp8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | gptj-99.9 | Server | Tokens/s | 8 | 20,706.7 | 2,588.3375 | available | fp8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | llama2-70b-99 | Offline | Tokens/s | 8 | 30,899.7 | 3,862.4625 | available | fp8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | llama2-70b-99 | Server | Tokens/s | 8 | 26,599.4 | 3,324.925 | available | fp8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | llama2-70b-99.9 | Offline | Tokens/s | 8 | 30,899.7 | 3,862.4625 | available | fp8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | llama2-70b-99.9 | Server | Tokens/s | 8 | 26,599.4 | 3,324.925 | available | fp8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | mixtral-8x7b | Offline | Tokens/s | 8 | 52,724.4 | 6,590.55 | available | fp8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | mixtral-8x7b | Server | Tokens/s | 8 | 53,299.3 | 6,662.4125 | available | fp8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | resnet | Offline | Samples/s | 8 | 715,761 | 89,470.125 | available | int8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | resnet | Server | Queries/s | 8 | 652,002 | 81,500.25 | available | int8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | retinanet | Offline | Samples/s | 8 | 14,476.3 | 1,809.5375 | available | int8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | retinanet | Server | Queries/s | 8 | 13,908.7 | 1,738.5875 | available | int8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | stable-diffusion-xl | Offline | Samples/s | 8 | 18.375 | 2.296875 | available | fp8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0010 | Cisco | NVIDIA H100-SXM-80GB | stable-diffusion-xl | Server | Queries/s | 8 | 17.7917 | 2.223963 | available | fp8 | Cisco UCS C885A M8 (8x H100-SXM-80GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | gptj-99 | Offline | Tokens/s | 8 | 21,557.8 | 2,694.725 | available | fp8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | gptj-99 | Server | Tokens/s | 8 | 21,813.4 | 2,726.675 | available | fp8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | gptj-99.9 | Offline | Tokens/s | 8 | 21,557.8 | 2,694.725 | available | fp8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | gptj-99.9 | Server | Tokens/s | 8 | 21,813.4 | 2,726.675 | available | fp8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | llama2-70b-99 | Offline | Tokens/s | 8 | 34,081.1 | 4,260.1375 | available | fp8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | llama2-70b-99 | Server | Tokens/s | 8 | 30,420.2 | 3,802.525 | available | fp8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | llama2-70b-99.9 | Offline | Tokens/s | 8 | 34,081.1 | 4,260.1375 | available | fp8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | llama2-70b-99.9 | Server | Tokens/s | 8 | 30,420.2 | 3,802.525 | available | fp8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | llama2-70b-interactive-99 | Offline | Tokens/s | 8 | 35,102.5 | 4,387.8125 | available | fp8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | llama2-70b-interactive-99 | Server | Tokens/s | 8 | 19,437.4 | 2,429.675 | available | fp8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | llama3.1-405b | Offline | Tokens/s | 8 | 573.573 | 71.696625 | available | fp8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | llama3.1-405b | Server | Tokens/s | 8 | 294.275 | 36.784375 | available | fp8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | resnet | Offline | Samples/s | 8 | 770,496 | 96,312 | available | int8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0011 | Cisco | NVIDIA H200-SXM-141GB | resnet | Server | Queries/s | 8 | 648,207 | 81,025.875 | available | int8 | Cisco UCS C885A M8 (8x H200-SXM-141GB, TensorRT) |
v5.0 | 5.0-0012 | Cisco | NVIDIA L40S | 3d-unet-99 | Offline | Samples/s | 2 | 7.7556 | 3.8778 | available | int8 | UCS C245 M8 (2x L40S, TensorRT) |
v5.0 | 5.0-0012 | Cisco | NVIDIA L40S | dlrm-v2-99 | Offline | Samples/s | 2 | 57,832.5 | 28,916.25 | available | int8 | UCS C245 M8 (2x L40S, TensorRT) |
v5.0 | 5.0-0012 | Cisco | NVIDIA L40S | dlrm-v2-99 | Server | Queries/s | 2 | 49,591.6 | 24,795.8 | available | int8 | UCS C245 M8 (2x L40S, TensorRT) |
v5.0 | 5.0-0012 | Cisco | NVIDIA L40S | stable-diffusion-xl | Offline | Samples/s | 2 | 1.43813 | 0.719065 | available | int8 | UCS C245 M8 (2x L40S, TensorRT) |
v5.0 | 5.0-0012 | Cisco | NVIDIA L40S | stable-diffusion-xl | Server | Queries/s | 2 | 1.26862 | 0.63431 | available | int8 | UCS C245 M8 (2x L40S, TensorRT) |
v5.0 | 5.0-0013 | Cisco | NVIDIA L40S | 3d-unet-99 | Offline | Samples/s | 2 | 7.68797 | 3.843985 | available | int8 | UCS X215 M8 (2x L40S, TensorRT) |
v5.0 | 5.0-0013 | Cisco | NVIDIA L40S | dlrm-v2-99 | Offline | Samples/s | 2 | 57,918.1 | 28,959.05 | available | int8 | UCS X215 M8 (2x L40S, TensorRT) |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.