benchmark.name
large_stringclasses 1
value | benchmark.version
large_stringclasses 2
values | datetime
large_stringclasses 1
value | metrics.accuracy
large_stringlengths 65
120
| metrics.result
float64 64
129k
| metrics.result_per_accelerator
float64 23.6
16.1k
| metrics.units
large_stringclasses 1
value | model.architecture
large_stringclasses 1
value | model.mlperf_name
large_stringclasses 8
values | model.name
large_stringclasses 19
values | model.number_of_parameters
float64 6
671
| model.weight_data_types
large_stringclasses 7
values | software.framework
large_stringclasses 28
values | software.framework.ai
large_stringclasses 0
values | software.framework.cuda
large_stringclasses 6
values | software.framework.cudnn
large_stringclasses 0
values | software.framework.jetpack
large_stringclasses 1
value | software.framework.pytorch
large_stringclasses 2
values | software.framework.rocm
large_stringclasses 5
values | software.framework.sima
large_stringclasses 0
values | software.framework.tensorrt
large_stringclasses 9
values | software.framework.vllm
large_stringclasses 4
values | software.operating_system
large_stringclasses 31
values | submission.availability
large_stringclasses 2
values | submission.debug_uid
large_stringlengths 16
16
| submission.division
large_stringclasses 2
values | submission.organization
large_stringclasses 24
values | submission.scenario
large_stringclasses 2
values | system.accelerator.count_per_node
int64 1
8
| system.accelerator.memory_capacity
float64 23.6
2.05k
⌀ | system.accelerator.memory_config
large_stringclasses 9
values | system.accelerator.name
large_stringclasses 20
values | system.accelerator.total_count
int64 1
72
| system.accelerator.vendor
large_stringclasses 2
values | system.cpu.caches
large_stringclasses 15
values | system.cpu.core_count
int64 4
384
⌀ | system.cpu.count_per_node
int64 1
4
| system.cpu.frequency
float64 2
4.8k
⌀ | system.cpu.model
large_stringclasses 56
values | system.cpu.vcpu_count
large_stringclasses 2
values | system.cpu.vendor
large_stringclasses 4
values | system.interconnect.accelerator
large_stringclasses 7
values | system.interconnect.accelerator_host
large_stringclasses 4
values | system.memory.capacity
float64 64
8.19k
⌀ | system.memory.configuration
large_stringclasses 66
values | system.name
large_stringclasses 109
values | system.number_of_nodes
int64 1
18
| system.type
large_stringclasses 2
values |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
mlperf-inference
|
4.1
| null |
ROUGE1: 45.3591 ROUGE2: 23.2517 ROUGEL: 30.3173 TOKENS_PER_SAMPLE: 145.9 gsm8k_accuracy: 73.1 mbxp_accuracy: 59.9
| 25,358.5 | 6,339.625 |
Tokens/s
|
LLM
|
mixtral-8x7b
|
mixtral-8x7b
| 47 |
fp16
|
TensorRT 10.2, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2
| null |
Ubuntu 20.04.4
|
available
|
56a93d420b2f4988
|
closed
|
Fujitsu
|
Server
| 4 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 4 |
NVIDIA
| null | 48 | 2 | null |
Intel(R) Xeon(R) Platinum 8468
| null |
Intel
|
NVLink-4
|
PCIe-5
| 1,024 |
32x 32GB DDR5
|
GX2560M7_H100_SXM_80GBx4 (4x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1017 ROUGE2: 20.1125 ROUGEL: 29.9751 GEN_LEN: 4114386
| 20,041.3 | 2,505.1625 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
5c2b7c9c5fef4ee5
|
closed
|
GigaComputing
|
Offline
| 8 | 141 |
HBM3
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB DDR5
|
GIGABYTE G593-SD1
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1004 ROUGE2: 20.1119 ROUGEL: 29.9737 GEN_LEN: 4114146
| 19,250.4 | 2,406.3 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
ff06eb82764a40c5
|
closed
|
GigaComputing
|
Server
| 8 | 141 |
HBM3
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB DDR5
|
GIGABYTE G593-SD1
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1017 ROUGE2: 20.1125 ROUGEL: 29.9751 GEN_LEN: 4114386
| 20,041.3 | 2,505.1625 |
Tokens/s
|
LLM
|
gptj-99.9
|
gptj-99.9
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
35b4b94e51be4de2
|
closed
|
GigaComputing
|
Offline
| 8 | 141 |
HBM3
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB DDR5
|
GIGABYTE G593-SD1
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1004 ROUGE2: 20.1119 ROUGEL: 29.9737 GEN_LEN: 4114146
| 19,250.4 | 2,406.3 |
Tokens/s
|
LLM
|
gptj-99.9
|
gptj-99.9
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
aaee26d8ea754b10
|
closed
|
GigaComputing
|
Server
| 8 | 141 |
HBM3
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB DDR5
|
GIGABYTE G593-SD1
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5884 ROUGE2: 22.1723 ROUGEL: 28.7905 TOKENS_PER_SAMPLE: 291.7
| 31,263.8 | 3,907.975 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
e7d36afea30846d6
|
closed
|
GigaComputing
|
Offline
| 8 | 141 |
HBM3
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB DDR5
|
GIGABYTE G593-SD1
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5891 ROUGE2: 22.1726 ROUGEL: 28.7906 TOKENS_PER_SAMPLE: 291.6
| 29,715.3 | 3,714.4125 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
78e88fbf1c2a44a2
|
closed
|
GigaComputing
|
Server
| 8 | 141 |
HBM3
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB DDR5
|
GIGABYTE G593-SD1
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5884 ROUGE2: 22.1723 ROUGEL: 28.7905 TOKENS_PER_SAMPLE: 291.7
| 31,263.8 | 3,907.975 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
6d3d8cbbfa4c48df
|
closed
|
GigaComputing
|
Offline
| 8 | 141 |
HBM3
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB DDR5
|
GIGABYTE G593-SD1
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5891 ROUGE2: 22.1726 ROUGEL: 28.7906 TOKENS_PER_SAMPLE: 291.6
| 29,715.3 | 3,714.4125 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
56c1065f1f494c9c
|
closed
|
GigaComputing
|
Server
| 8 | 141 |
HBM3
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB DDR5
|
GIGABYTE G593-SD1
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 45.3568 ROUGE2: 23.2496 ROUGEL: 30.3171 TOKENS_PER_SAMPLE: 145.9 gsm8k_accuracy: 73.1 mbxp_accuracy: 59.88
| 59,334.8 | 7,416.85 |
Tokens/s
|
LLM
|
mixtral-8x7b
|
mixtral-8x7b
| 47 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
38470b18dc7e44cc
|
closed
|
GigaComputing
|
Offline
| 8 | 141 |
HBM3
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB DDR5
|
GIGABYTE G593-SD1
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 45.3551 ROUGE2: 23.245 ROUGEL: 30.3156 TOKENS_PER_SAMPLE: 146.0 gsm8k_accuracy: 73.1 mbxp_accuracy: 59.88
| 57,174.1 | 7,146.7625 |
Tokens/s
|
LLM
|
mixtral-8x7b
|
mixtral-8x7b
| 47 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
cc3317c06f794c13
|
closed
|
GigaComputing
|
Server
| 8 | 141 |
HBM3
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB DDR5
|
GIGABYTE G593-SD1
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6156 ROUGE2: 22.1792 ROUGEL: 28.8002 TOKENS_PER_SAMPLE: 290.1
| 24,133.7 | 3,016.7125 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
5772acdace8c444e
|
closed
|
Google
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | 2.7 |
Intel(R) Xeon(R) Platinum 8481C
| null |
Intel
|
NVLink-4
|
PCIe
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6146 ROUGE2: 22.1791 ROUGEL: 28.7999 TOKENS_PER_SAMPLE: 290.1
| 21,588.3 | 2,698.5375 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
21d74a9a24b845c5
|
closed
|
Google
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | 2.7 |
Intel(R) Xeon(R) Platinum 8481C
| null |
Intel
|
NVLink-4
|
PCIe
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6156 ROUGE2: 22.1792 ROUGEL: 28.8002 TOKENS_PER_SAMPLE: 290.1
| 24,133.7 | 3,016.7125 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
41f703faa8e6486c
|
closed
|
Google
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | 2.7 |
Intel(R) Xeon(R) Platinum 8481C
| null |
Intel
|
NVLink-4
|
PCIe
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6146 ROUGE2: 22.1791 ROUGEL: 28.7999 TOKENS_PER_SAMPLE: 290.1
| 21,588.3 | 2,698.5375 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
5a0ab3e8c9534831
|
closed
|
Google
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | 2.7 |
Intel(R) Xeon(R) Platinum 8481C
| null |
Intel
|
NVLink-4
|
PCIe
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 45.3671 ROUGE2: 23.2507 ROUGEL: 30.3151 TOKENS_PER_SAMPLE: 145.9 gsm8k_accuracy: 73.1 mbxp_accuracy: 59.88
| 52,524.4 | 6,565.55 |
Tokens/s
|
LLM
|
mixtral-8x7b
|
mixtral-8x7b
| 47 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
e7a9d4a3f9cf4f34
|
closed
|
Google
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | 2.7 |
Intel(R) Xeon(R) Platinum 8481C
| null |
Intel
|
NVLink-4
|
PCIe
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 45.3561 ROUGE2: 23.2477 ROUGEL: 30.3155 TOKENS_PER_SAMPLE: 145.9 gsm8k_accuracy: 73.08 mbxp_accuracy: 59.88
| 50,798.8 | 6,349.85 |
Tokens/s
|
LLM
|
mixtral-8x7b
|
mixtral-8x7b
| 47 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
cc753adf51ee430c
|
closed
|
Google
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | 2.7 |
Intel(R) Xeon(R) Platinum 8481C
| null |
Intel
|
NVLink-4
|
PCIe
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1017 ROUGE2: 20.1125 ROUGEL: 29.9751 GEN_LEN: 4114386
| 2,632.76 | 2,632.76 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
preview
|
c14a3d84c0774058
|
closed
|
HPE
|
Offline
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
HPE ProLiant Compute DL384 Gen12 (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1002 ROUGE2: 20.1114 ROUGEL: 29.9742 GEN_LEN: 4114259
| 2,512.89 | 2,512.89 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
preview
|
187a784531de4493
|
closed
|
HPE
|
Server
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
HPE ProLiant Compute DL384 Gen12 (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1017 ROUGE2: 20.1125 ROUGEL: 29.9751 GEN_LEN: 4114386
| 2,632.76 | 2,632.76 |
Tokens/s
|
LLM
|
gptj-99.9
|
gptj-99.9
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
preview
|
29a6c1e05c0f4014
|
closed
|
HPE
|
Offline
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
HPE ProLiant Compute DL384 Gen12 (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1002 ROUGE2: 20.1114 ROUGEL: 29.9742 GEN_LEN: 4114259
| 2,512.89 | 2,512.89 |
Tokens/s
|
LLM
|
gptj-99.9
|
gptj-99.9
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
preview
|
f5b8a0e9da3249cb
|
closed
|
HPE
|
Server
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
HPE ProLiant Compute DL384 Gen12 (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5631 ROUGE2: 22.1578 ROUGEL: 28.774 TOKENS_PER_SAMPLE: 291.2
| 4,084.3 | 4,084.3 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
preview
|
a4aac70681614b69
|
closed
|
HPE
|
Offline
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
HPE ProLiant Compute DL384 Gen12 (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.563 ROUGE2: 22.1577 ROUGEL: 28.7741 TOKENS_PER_SAMPLE: 291.3
| 3,884.16 | 3,884.16 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
preview
|
44fb91d79cfa4566
|
closed
|
HPE
|
Server
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
HPE ProLiant Compute DL384 Gen12 (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5631 ROUGE2: 22.1578 ROUGEL: 28.774 TOKENS_PER_SAMPLE: 291.2
| 4,084.3 | 4,084.3 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
preview
|
5401414114074d9d
|
closed
|
HPE
|
Offline
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
HPE ProLiant Compute DL384 Gen12 (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.563 ROUGE2: 22.1577 ROUGEL: 28.7741 TOKENS_PER_SAMPLE: 291.3
| 3,884.16 | 3,884.16 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
preview
|
4da84d7e93844b4c
|
closed
|
HPE
|
Server
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
HPE ProLiant Compute DL384 Gen12 (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 45.3565 ROUGE2: 23.2488 ROUGEL: 30.3157 TOKENS_PER_SAMPLE: 145.9 gsm8k_accuracy: 73.1 mbxp_accuracy: 59.9
| 8,063.02 | 8,063.02 |
Tokens/s
|
LLM
|
mixtral-8x7b
|
mixtral-8x7b
| 47 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
preview
|
64bf3ffb97a64254
|
closed
|
HPE
|
Offline
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
HPE ProLiant Compute DL384 Gen12 (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 45.3563 ROUGE2: 23.2486 ROUGEL: 30.3162 TOKENS_PER_SAMPLE: 145.9 gsm8k_accuracy: 73.1 mbxp_accuracy: 59.94
| 7,450.72 | 7,450.72 |
Tokens/s
|
LLM
|
mixtral-8x7b
|
mixtral-8x7b
| 47 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
preview
|
3165e4ad4fb44efe
|
closed
|
HPE
|
Server
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
HPE ProLiant Compute DL384 Gen12 (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.0343 ROUGE2: 20.0912 ROUGEL: 29.9254 GEN_LEN: 4102194
| 19,751.9 | 2,468.9875 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
22.04.4 LTS
|
available
|
f1a35ef201f74997
|
closed
|
HPE
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
|
L1d: 4.5 MiB (96 instances), L1i: 3 MiB (96 instances), L2: 192 MiB (96 instances), L3: 210 MiB (2 instances)
| 48 | 2 | 3.8 |
Intel(R) Xeon(R) Platinum 8468
| null |
Intel
|
NVLink-4
| null | 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
HPE Cray XD670 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.0345 ROUGE2: 20.0914 ROUGEL: 29.9253 GEN_LEN: 4102160
| 19,502.6 | 2,437.825 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
22.04.4 LTS
|
available
|
94fe52da4671446c
|
closed
|
HPE
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
|
L1d: 4.5 MiB (96 instances), L1i: 3 MiB (96 instances), L2: 192 MiB (96 instances), L3: 210 MiB (2 instances)
| 48 | 2 | 3.8 |
Intel(R) Xeon(R) Platinum 8468
| null |
Intel
|
NVLink-4
| null | 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
HPE Cray XD670 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.535 ROUGE2: 22.0991 ROUGEL: 28.7319 TOKENS_PER_SAMPLE: 290.9
| 24,528.3 | 3,066.0375 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
22.04.4 LTS
|
available
|
6445c436a3f04274
|
closed
|
HPE
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
|
L1d: 4.5 MiB (96 instances), L1i: 3 MiB (96 instances), L2: 192 MiB (96 instances), L3: 210 MiB (2 instances)
| 48 | 2 | 3.8 |
Intel(R) Xeon(R) Platinum 8468
| null |
Intel
|
NVLink-4
| null | 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
HPE Cray XD670 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5353 ROUGE2: 22.0997 ROUGEL: 28.7319 TOKENS_PER_SAMPLE: 290.9
| 23,132.8 | 2,891.6 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
22.04.4 LTS
|
available
|
a035dcdac0554817
|
closed
|
HPE
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
|
L1d: 4.5 MiB (96 instances), L1i: 3 MiB (96 instances), L2: 192 MiB (96 instances), L3: 210 MiB (2 instances)
| 48 | 2 | 3.8 |
Intel(R) Xeon(R) Platinum 8468
| null |
Intel
|
NVLink-4
| null | 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
HPE Cray XD670 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.535 ROUGE2: 22.0991 ROUGEL: 28.7319 TOKENS_PER_SAMPLE: 290.9
| 24,424.7 | 3,053.0875 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
22.04.4 LTS
|
available
|
fba4777a50f149ad
|
closed
|
HPE
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
|
L1d: 4.5 MiB (96 instances), L1i: 3 MiB (96 instances), L2: 192 MiB (96 instances), L3: 210 MiB (2 instances)
| 48 | 2 | 3.8 |
Intel(R) Xeon(R) Platinum 8468
| null |
Intel
|
NVLink-4
| null | 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
HPE Cray XD670 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5353 ROUGE2: 22.0997 ROUGEL: 28.732 TOKENS_PER_SAMPLE: 290.9
| 23,144 | 2,893 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
22.04.4 LTS
|
available
|
2e3f6e8e41814a00
|
closed
|
HPE
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
|
L1d: 4.5 MiB (96 instances), L1i: 3 MiB (96 instances), L2: 192 MiB (96 instances), L3: 210 MiB (2 instances)
| 48 | 2 | 3.8 |
Intel(R) Xeon(R) Platinum 8468
| null |
Intel
|
NVLink-4
| null | 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
HPE Cray XD670 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 45.3566 ROUGE2: 23.2487 ROUGEL: 30.3174 TOKENS_PER_SAMPLE: 146.0 gsm8k_accuracy: 73.1 mbxp_accuracy: 59.9
| 52,395.1 | 6,549.3875 |
Tokens/s
|
LLM
|
mixtral-8x7b
|
mixtral-8x7b
| 47 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
22.04.4 LTS
|
available
|
1f82570acc854582
|
closed
|
HPE
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
|
L1d: 4.5 MiB (96 instances), L1i: 3 MiB (96 instances), L2: 192 MiB (96 instances), L3: 210 MiB (2 instances)
| 48 | 2 | 3.8 |
Intel(R) Xeon(R) Platinum 8468
| null |
Intel
|
NVLink-4
| null | 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
HPE Cray XD670 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 45.3565 ROUGE2: 23.249 ROUGEL: 30.3137 TOKENS_PER_SAMPLE: 146.0 gsm8k_accuracy: 73.1 mbxp_accuracy: 59.9
| 50,798.5 | 6,349.8125 |
Tokens/s
|
LLM
|
mixtral-8x7b
|
mixtral-8x7b
| 47 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
22.04.4 LTS
|
available
|
2fe89086851f4953
|
closed
|
HPE
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
|
L1d: 4.5 MiB (96 instances), L1i: 3 MiB (96 instances), L2: 192 MiB (96 instances), L3: 210 MiB (2 instances)
| 48 | 2 | 3.8 |
Intel(R) Xeon(R) Platinum 8468
| null |
Intel
|
NVLink-4
| null | 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
HPE Cray XD670 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6113 ROUGE2: 22.1451 ROUGEL: 28.7779 TOKENS_PER_SAMPLE: 290.3
| 8,858.78 | 2,214.695 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4 LTS
|
available
|
bebca7d994ef4ed3
|
closed
|
HPE
|
Offline
| 4 | 94 |
HBM3
|
NVIDIA H100-NVL-94GB
| 4 |
NVIDIA
|
L1d: 3 MiB, L1i: 2 MiB, L2: 128 MiB (64 instances), L3: 320 MiB (2 instances)
| 32 | 2 | 4 |
INTEL(R) XEON(R) GOLD 6530
| null |
Intel
|
NVLink-4
| null | 1,024 |
16x 64GB
|
HPE ProLiant DL380a Gen11 (4x H100-NVL-94GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6118 ROUGE2: 22.1456 ROUGEL: 28.7778 TOKENS_PER_SAMPLE: 290.3
| 6,927.8 | 1,731.95 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4 LTS
|
available
|
9d4f6b34ff2e4c0f
|
closed
|
HPE
|
Server
| 4 | 94 |
HBM3
|
NVIDIA H100-NVL-94GB
| 4 |
NVIDIA
|
L1d: 3 MiB, L1i: 2 MiB, L2: 128 MiB (64 instances), L3: 320 MiB (2 instances)
| 32 | 2 | 4 |
INTEL(R) XEON(R) GOLD 6530
| null |
Intel
|
NVLink-4
| null | 1,024 |
16x 64GB
|
HPE ProLiant DL380a Gen11 (4x H100-NVL-94GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6113 ROUGE2: 22.1451 ROUGEL: 28.7779 TOKENS_PER_SAMPLE: 290.3
| 8,858.78 | 2,214.695 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4 LTS
|
available
|
a63bc04d461b413b
|
closed
|
HPE
|
Offline
| 4 | 94 |
HBM3
|
NVIDIA H100-NVL-94GB
| 4 |
NVIDIA
|
L1d: 3 MiB, L1i: 2 MiB, L2: 128 MiB (64 instances), L3: 320 MiB (2 instances)
| 32 | 2 | 4 |
INTEL(R) XEON(R) GOLD 6530
| null |
Intel
|
NVLink-4
| null | 1,024 |
16x 64GB
|
HPE ProLiant DL380a Gen11 (4x H100-NVL-94GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6118 ROUGE2: 22.1456 ROUGEL: 28.7778 TOKENS_PER_SAMPLE: 290.3
| 6,927.8 | 1,731.95 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4 LTS
|
available
|
d6fda6fd06c14da9
|
closed
|
HPE
|
Server
| 4 | 94 |
HBM3
|
NVIDIA H100-NVL-94GB
| 4 |
NVIDIA
|
L1d: 3 MiB, L1i: 2 MiB, L2: 128 MiB (64 instances), L3: 320 MiB (2 instances)
| 32 | 2 | 4 |
INTEL(R) XEON(R) GOLD 6530
| null |
Intel
|
NVLink-4
| null | 1,024 |
16x 64GB
|
HPE ProLiant DL380a Gen11 (4x H100-NVL-94GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6047 ROUGE2: 22.1268 ROUGEL: 28.7705 TOKENS_PER_SAMPLE: 291.9
| 41,672.4 | 2,604.525 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 9.3.0, CUDA 12.2
| null |
12.2
| null | null | null | null | null |
9.3.0
| null |
Ubuntu 20.04.4
|
available
|
b03a31db50504cb7
|
closed
|
JuniperNetworks
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 16 |
NVIDIA
| null | 112 | 2 | null |
AMD EPYC 7763 64-Core Processor
| null |
AMD
|
NVLink
|
PCIe
| 2,048 |
32 slots / 64 GB each/ Total = 2 TB /DDR4 3200 MHz
|
2x8xH100-SXM-80GB
| 2 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6052 ROUGE2: 22.1275 ROUGEL: 28.7709 TOKENS_PER_SAMPLE: 291.9
| 41,091.6 | 2,568.225 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 9.3.0, CUDA 12.2
| null |
12.2
| null | null | null | null | null |
9.3.0
| null |
Ubuntu 20.04.4
|
available
|
13d079231fd54dfa
|
closed
|
JuniperNetworks
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 16 |
NVIDIA
| null | 112 | 2 | null |
AMD EPYC 7763 64-Core Processor
| null |
AMD
|
NVLink
|
PCIe
| 2,048 |
32 slots / 64 GB each/ Total = 2 TB /DDR4 3200 MHz
|
2x8xH100-SXM-80GB
| 2 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6047 ROUGE2: 22.1268 ROUGEL: 28.7705 TOKENS_PER_SAMPLE: 291.9
| 41,672.4 | 2,604.525 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 9.3.0, CUDA 12.2
| null |
12.2
| null | null | null | null | null |
9.3.0
| null |
Ubuntu 20.04.4
|
available
|
3fa08230362d4ae6
|
closed
|
JuniperNetworks
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 16 |
NVIDIA
| null | 112 | 2 | null |
AMD EPYC 7763 64-Core Processor
| null |
AMD
|
NVLink
|
PCIe
| 2,048 |
32 slots / 64 GB each/ Total = 2 TB /DDR4 3200 MHz
|
2x8xH100-SXM-80GB
| 2 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6052 ROUGE2: 22.1275 ROUGEL: 28.7709 TOKENS_PER_SAMPLE: 291.9
| 41,091.6 | 2,568.225 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 9.3.0, CUDA 12.2
| null |
12.2
| null | null | null | null | null |
9.3.0
| null |
Ubuntu 20.04.4
|
available
|
ba6c84f08d2645a9
|
closed
|
JuniperNetworks
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 16 |
NVIDIA
| null | 112 | 2 | null |
AMD EPYC 7763 64-Core Processor
| null |
AMD
|
NVLink
|
PCIe
| 2,048 |
32 slots / 64 GB each/ Total = 2 TB /DDR4 3200 MHz
|
2x8xH100-SXM-80GB
| 2 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6048 ROUGE2: 22.1273 ROUGEL: 28.7697 TOKENS_PER_SAMPLE: 291.9
| 82,749.6 | 2,585.925 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 9.3.0, CUDA 12.2
| null |
12.2
| null | null | null | null | null |
9.3.0
| null |
Ubuntu 20.04.4
|
available
|
2b9a57fa390c4521
|
closed
|
JuniperNetworks
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 32 |
NVIDIA
| null | 112 | 2 | null |
AMD EPYC 7763 64-Core Processor
| null |
AMD
|
NVLink
|
PCIe
| 2,048 |
32 slots / 64 GB each/ Total = 2 TB /DDR4 3200 MHz
|
4x8xH100-SXM-80GB
| 4 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6056 ROUGE2: 22.1274 ROUGEL: 28.7709 TOKENS_PER_SAMPLE: 291.9
| 82,273.2 | 2,571.0375 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 9.3.0, CUDA 12.2
| null |
12.2
| null | null | null | null | null |
9.3.0
| null |
Ubuntu 20.04.4
|
available
|
203986133b5c42ce
|
closed
|
JuniperNetworks
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 32 |
NVIDIA
| null | 112 | 2 | null |
AMD EPYC 7763 64-Core Processor
| null |
AMD
|
NVLink
|
PCIe
| 2,048 |
32 slots / 64 GB each/ Total = 2 TB /DDR4 3200 MHz
|
4x8xH100-SXM-80GB
| 4 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6048 ROUGE2: 22.1273 ROUGEL: 28.7697 TOKENS_PER_SAMPLE: 291.9
| 82,749.6 | 2,585.925 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 9.3.0, CUDA 12.2
| null |
12.2
| null | null | null | null | null |
9.3.0
| null |
Ubuntu 20.04.4
|
available
|
1f3b63740e144aba
|
closed
|
JuniperNetworks
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 32 |
NVIDIA
| null | 112 | 2 | null |
AMD EPYC 7763 64-Core Processor
| null |
AMD
|
NVLink
|
PCIe
| 2,048 |
32 slots / 64 GB each/ Total = 2 TB /DDR4 3200 MHz
|
4x8xH100-SXM-80GB
| 4 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6056 ROUGE2: 22.1274 ROUGEL: 28.7709 TOKENS_PER_SAMPLE: 291.9
| 82,273.2 | 2,571.0375 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 9.3.0, CUDA 12.2
| null |
12.2
| null | null | null | null | null |
9.3.0
| null |
Ubuntu 20.04.4
|
available
|
f0d176ad37e34749
|
closed
|
JuniperNetworks
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 32 |
NVIDIA
| null | 112 | 2 | null |
AMD EPYC 7763 64-Core Processor
| null |
AMD
|
NVLink
|
PCIe
| 2,048 |
32 slots / 64 GB each/ Total = 2 TB /DDR4 3200 MHz
|
4x8xH100-SXM-80GB
| 4 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.0773 ROUGE2: 20.0872 ROUGEL: 29.9583 GEN_LEN: 4107356
| 19,859.2 | 2,482.4 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
828a2c1a8930482f
|
closed
|
Lenovo
|
Offline
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
AMD EPYC 9454
| null |
AMD
|
NVLink-4
|
PCIe-5
| 1,536 |
24x 64GB HMCG94MEBRA121N
|
ThinkSystem SR685a V3(8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.0786 ROUGE2: 20.0893 ROUGEL: 29.9596 GEN_LEN: 4107309
| 19,716 | 2,464.5 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
ae403719735f41ee
|
closed
|
Lenovo
|
Server
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
AMD EPYC 9454
| null |
AMD
|
NVLink-4
|
PCIe-5
| 1,536 |
24x 64GB HMCG94MEBRA121N
|
ThinkSystem SR685a V3(8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6249 ROUGE2: 22.1876 ROUGEL: 28.8076 TOKENS_PER_SAMPLE: 290.9
| 31,917 | 3,989.625 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
6dd468249e8646b9
|
closed
|
Lenovo
|
Offline
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
AMD EPYC 9454
| null |
AMD
|
NVLink-4
|
PCIe-5
| 1,536 |
24x 64GB HMCG94MEBRA121N
|
ThinkSystem SR685a V3(8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6256 ROUGE2: 22.1883 ROUGEL: 28.808 TOKENS_PER_SAMPLE: 290.9
| 30,068.5 | 3,758.5625 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
bd6d4b0912a14d2b
|
closed
|
Lenovo
|
Server
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
AMD EPYC 9454
| null |
AMD
|
NVLink-4
|
PCIe-5
| 1,536 |
24x 64GB HMCG94MEBRA121N
|
ThinkSystem SR685a V3(8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6249 ROUGE2: 22.1876 ROUGEL: 28.8076 TOKENS_PER_SAMPLE: 290.9
| 31,917 | 3,989.625 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
c30a2b878c3e40c2
|
closed
|
Lenovo
|
Offline
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
AMD EPYC 9454
| null |
AMD
|
NVLink-4
|
PCIe-5
| 1,536 |
24x 64GB HMCG94MEBRA121N
|
ThinkSystem SR685a V3(8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.6256 ROUGE2: 22.1883 ROUGEL: 28.808 TOKENS_PER_SAMPLE: 290.9
| 30,068.5 | 3,758.5625 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
308c7beb01cc48eb
|
closed
|
Lenovo
|
Server
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
AMD EPYC 9454
| null |
AMD
|
NVLink-4
|
PCIe-5
| 1,536 |
24x 64GB HMCG94MEBRA121N
|
ThinkSystem SR685a V3(8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.0266 ROUGE2: 20.1037 ROUGEL: 29.9635 GEN_LEN: 4086524
| 1,718.33 | 859.165 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 9.3.0, CUDA 12.3
| null |
12.3
| null | null | null | null | null |
9.3.0
| null |
Ubuntu 22.04.3
|
available
|
6e13e5328dce458c
|
closed
|
Lenovo
|
Offline
| 2 | 48 |
GDDR6
|
NVIDIA L40S
| 2 |
NVIDIA
| null | 16 | 1 | null |
AMD EPYC 8124P 16-Core Processor
| null |
AMD
| null | null | 128 |
4x 32GB M321R4GA3BB6-CQKDS
|
ThinkEdge SE455 V3 (2x NVIDIA L40S, TensorRT)
| 1 |
edge
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.102 ROUGE2: 20.1124 ROUGEL: 29.9753 GEN_LEN: 4114427
| 20,552.1 | 2,569.0125 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
27b24f7ee627439d
|
closed
|
Lenovo
|
Offline
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
Intel(R) Xeon(R) Platinum 8568Y+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
ThinkSystem SR680a V3 (8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.101 ROUGE2: 20.1122 ROUGEL: 29.9738 GEN_LEN: 4114231
| 19,785 | 2,473.125 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
b5899af5e0094c7c
|
closed
|
Lenovo
|
Server
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
Intel(R) Xeon(R) Platinum 8568Y+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
ThinkSystem SR680a V3 (8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.102 ROUGE2: 20.1124 ROUGEL: 29.9753 GEN_LEN: 4114427
| 20,552.1 | 2,569.0125 |
Tokens/s
|
LLM
|
gptj-99.9
|
gptj-99.9
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
7a4405625fb44984
|
closed
|
Lenovo
|
Offline
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
Intel(R) Xeon(R) Platinum 8568Y+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
ThinkSystem SR680a V3 (8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.101 ROUGE2: 20.1122 ROUGEL: 29.9738 GEN_LEN: 4114231
| 19,785 | 2,473.125 |
Tokens/s
|
LLM
|
gptj-99.9
|
gptj-99.9
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
9f522fe48b5a404f
|
closed
|
Lenovo
|
Server
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
Intel(R) Xeon(R) Platinum 8568Y+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
ThinkSystem SR680a V3 (8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5884 ROUGE2: 22.1723 ROUGEL: 28.7905 TOKENS_PER_SAMPLE: 291.7
| 31,973.7 | 3,996.7125 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
0cdf2ba561884829
|
closed
|
Lenovo
|
Offline
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
Intel(R) Xeon(R) Platinum 8568Y+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
ThinkSystem SR680a V3 (8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5866 ROUGE2: 22.1706 ROUGEL: 28.789 TOKENS_PER_SAMPLE: 291.7
| 30,438.7 | 3,804.8375 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
db8a37b32727429b
|
closed
|
Lenovo
|
Server
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
Intel(R) Xeon(R) Platinum 8568Y+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
ThinkSystem SR680a V3 (8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5884 ROUGE2: 22.1723 ROUGEL: 28.7905 TOKENS_PER_SAMPLE: 291.7
| 31,973.7 | 3,996.7125 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
bdc320bb10c44e32
|
closed
|
Lenovo
|
Offline
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
Intel(R) Xeon(R) Platinum 8568Y+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
ThinkSystem SR680a V3 (8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5866 ROUGE2: 22.1706 ROUGEL: 28.789 TOKENS_PER_SAMPLE: 291.7
| 30,438.7 | 3,804.8375 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
42ec4d7de01c4c34
|
closed
|
Lenovo
|
Server
| 8 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB
| 8 |
NVIDIA
| null | 48 | 2 | null |
Intel(R) Xeon(R) Platinum 8568Y+
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
ThinkSystem SR680a V3 (8x H200-SXM-141GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.0516 ROUGE2: 20.0942 ROUGEL: 29.9288 GEN_LEN: 4111570
| 2,593.65 | 864.55 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
c3d81304d4354b33
|
closed
|
Lenovo
|
Offline
| 3 | 48 |
GDDR6
|
NVIDIA L40S
| 3 |
NVIDIA
| null | 32 | 2 | null |
Intel(R) Xeon(R) Gold 6438N
| null |
Intel
| null | null | 512 |
16x 32GB M321R4GA3BB6-CQKDS
|
ThinkSystem SR650 V3 (3x NVIDIA L40S, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.0516 ROUGE2: 20.0939 ROUGEL: 29.9288 GEN_LEN: 4111576
| 2,475.22 | 825.073333 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
eacbcb7d34514c72
|
closed
|
Lenovo
|
Server
| 3 | 48 |
GDDR6
|
NVIDIA L40S
| 3 |
NVIDIA
| null | 32 | 2 | null |
Intel(R) Xeon(R) Gold 6438N
| null |
Intel
| null | null | 512 |
16x 32GB M321R4GA3BB6-CQKDS
|
ThinkSystem SR650 V3 (3x NVIDIA L40S, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.0394 ROUGE2: 20.0943 ROUGEL: 29.9329 GEN_LEN: 4100159
| 14,073.6 | 1,759.2 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
e646525864594aa9
|
closed
|
Lenovo
|
Offline
| 8 | 94 |
HBM3
|
NVIDIA H100-NVL-94GB
| 8 |
NVIDIA
| null | 84 | 2 | null |
AMD EPYC 9634 84-Core Processor
| null |
AMD
|
NVLink-4
|
PCIe
| 3,072 |
24x 128GB HMCT04MEERA131N
|
ThinkSystem SR675 V3 (8x H100-NVL-94GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.0404 ROUGE2: 20.0949 ROUGEL: 29.9329 GEN_LEN: 4100496
| 13,083.5 | 1,635.4375 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
c84c41f3ee884012
|
closed
|
Lenovo
|
Server
| 8 | 94 |
HBM3
|
NVIDIA H100-NVL-94GB
| 8 |
NVIDIA
| null | 84 | 2 | null |
AMD EPYC 9634 84-Core Processor
| null |
AMD
|
NVLink-4
|
PCIe
| 3,072 |
24x 128GB HMCT04MEERA131N
|
ThinkSystem SR675 V3 (8x H100-NVL-94GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5999 ROUGE2: 22.1861 ROUGEL: 28.805 TOKENS_PER_SAMPLE: 291.4
| 15,875.8 | 1,984.475 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
bf52f318abf8491c
|
closed
|
Lenovo
|
Offline
| 8 | 94 |
HBM3
|
NVIDIA H100-NVL-94GB
| 8 |
NVIDIA
| null | 84 | 2 | null |
AMD EPYC 9634 84-Core Processor
| null |
AMD
|
NVLink-4
|
PCIe
| 3,072 |
24x 128GB HMCT04MEERA131N
|
ThinkSystem SR675 V3 (8x H100-NVL-94GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5887 ROUGE2: 22.1721 ROUGEL: 28.7905 TOKENS_PER_SAMPLE: 291.7
| 13,074.7 | 1,634.3375 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
5d57867037354137
|
closed
|
Lenovo
|
Server
| 8 | 94 |
HBM3
|
NVIDIA H100-NVL-94GB
| 8 |
NVIDIA
| null | 84 | 2 | null |
AMD EPYC 9634 84-Core Processor
| null |
AMD
|
NVLink-4
|
PCIe
| 3,072 |
24x 128GB HMCT04MEERA131N
|
ThinkSystem SR675 V3 (8x H100-NVL-94GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5999 ROUGE2: 22.1861 ROUGEL: 28.805 TOKENS_PER_SAMPLE: 291.4
| 15,875.8 | 1,984.475 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
3cad48b43768402e
|
closed
|
Lenovo
|
Offline
| 8 | 94 |
HBM3
|
NVIDIA H100-NVL-94GB
| 8 |
NVIDIA
| null | 84 | 2 | null |
AMD EPYC 9634 84-Core Processor
| null |
AMD
|
NVLink-4
|
PCIe
| 3,072 |
24x 128GB HMCT04MEERA131N
|
ThinkSystem SR675 V3 (8x H100-NVL-94GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5887 ROUGE2: 22.1721 ROUGEL: 28.7905 TOKENS_PER_SAMPLE: 291.7
| 13,074.7 | 1,634.3375 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.5
| null |
12.5
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
b37b755c61e449c6
|
closed
|
Lenovo
|
Server
| 8 | 94 |
HBM3
|
NVIDIA H100-NVL-94GB
| 8 |
NVIDIA
| null | 84 | 2 | null |
AMD EPYC 9634 84-Core Processor
| null |
AMD
|
NVLink-4
|
PCIe
| 3,072 |
24x 128GB HMCT04MEERA131N
|
ThinkSystem SR675 V3 (8x H100-NVL-94GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.3329 ROUGE2: 22.0197 ROUGEL: 28.598 TOKENS_PER_SAMPLE: 300.5
| 948.198 | 118.52475 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp8
|
vLLM 0.5.2
| null | null | null | null | null | null | null | null |
0.5.2
|
Ubuntu 22.04 (linux-5.15.0-94-generic-glibc2.35)
|
available
|
8299e3dd5e014aff
|
closed
|
NeuralMagic
|
Offline
| 8 | 48 |
GDDR6
|
NVIDIA L40S
| 8 |
NVIDIA
|
L1d cache: 1.3 MiB (40 instances), L1i cache: 1.3 MiB (40 instances), L2 cache: 40 MiB (40 instances), L3 cache: 160 MiB (5 instances)
| 4 | 2 | null |
AMD EPYC 9254 24-Core Processor
| null |
AMD
| null |
PCIe
| null |
undefined
|
Crusoe Cloud L40S (8x L40S PCIe, vLLM, FP8)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.4766 ROUGE2: 22.0865 ROUGEL: 28.6953 TOKENS_PER_SAMPLE: 295.9
| 592.265 | 74.033125 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp8
|
vLLM 0.5.2
| null | null | null | null | null | null | null | null |
0.5.2
|
Ubuntu 22.04 (linux-5.15.0-94-generic-glibc2.35)
|
available
|
994bbea7f1a341c7
|
closed
|
NeuralMagic
|
Server
| 8 | 48 |
GDDR6
|
NVIDIA L40S
| 8 |
NVIDIA
|
L1d cache: 1.3 MiB (40 instances), L1i cache: 1.3 MiB (40 instances), L2 cache: 40 MiB (40 instances), L3 cache: 160 MiB (5 instances)
| 4 | 2 | null |
AMD EPYC 9254 24-Core Processor
| null |
AMD
| null |
PCIe
| null |
undefined
|
Crusoe Cloud L40S (8x L40S PCIe, vLLM, FP8)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.7986 ROUGE2: 22.3524 ROUGEL: 29.161 TOKENS_PER_SAMPLE: 274.0
| 11,264.4 | 11,264.4 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.1.0, CUDA 12.7
| null |
12.7
| null | null | null | null | null |
10.1.0
| null |
Ubuntu 22.04.4
|
preview
|
5875c39c70854982
|
closed
|
NVIDIA
|
Offline
| 1 | 180 |
HBM3e
|
NVIDIA B200-SXM-180GB
| 1 |
NVIDIA
| null | 12 | 1 | null |
Intel(R) Xeon(R) Silver 4410Y
| null |
Intel
| null |
PCIe-5
| 408 |
6x 64GB MTC40F2046S1RC48BA1
|
NVIDIA B200 (1x B200-SXM-180GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.8023 ROUGE2: 22.355 ROUGEL: 29.1623 TOKENS_PER_SAMPLE: 273.9
| 10,755.6 | 10,755.6 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.1.0, CUDA 12.7
| null |
12.7
| null | null | null | null | null |
10.1.0
| null |
Ubuntu 22.04.4
|
preview
|
f3086f2193bb40a4
|
closed
|
NVIDIA
|
Server
| 1 | 180 |
HBM3e
|
NVIDIA B200-SXM-180GB
| 1 |
NVIDIA
| null | 12 | 1 | null |
Intel(R) Xeon(R) Silver 4410Y
| null |
Intel
| null |
PCIe-5
| 408 |
6x 64GB MTC40F2046S1RC48BA1
|
NVIDIA B200 (1x B200-SXM-180GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.7986 ROUGE2: 22.3524 ROUGEL: 29.161 TOKENS_PER_SAMPLE: 274.0
| 11,264.4 | 11,264.4 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.1.0, CUDA 12.7
| null |
12.7
| null | null | null | null | null |
10.1.0
| null |
Ubuntu 22.04.4
|
preview
|
c2da4cca7f8c4a8d
|
closed
|
NVIDIA
|
Offline
| 1 | 180 |
HBM3e
|
NVIDIA B200-SXM-180GB
| 1 |
NVIDIA
| null | 12 | 1 | null |
Intel(R) Xeon(R) Silver 4410Y
| null |
Intel
| null |
PCIe-5
| 408 |
6x 64GB MTC40F2046S1RC48BA1
|
NVIDIA B200 (1x B200-SXM-180GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.8023 ROUGE2: 22.355 ROUGEL: 29.1623 TOKENS_PER_SAMPLE: 273.9
| 10,755.6 | 10,755.6 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.1.0, CUDA 12.7
| null |
12.7
| null | null | null | null | null |
10.1.0
| null |
Ubuntu 22.04.4
|
preview
|
7c61a69a4ce94f70
|
closed
|
NVIDIA
|
Server
| 1 | 180 |
HBM3e
|
NVIDIA B200-SXM-180GB
| 1 |
NVIDIA
| null | 12 | 1 | null |
Intel(R) Xeon(R) Silver 4410Y
| null |
Intel
| null |
PCIe-5
| 408 |
6x 64GB MTC40F2046S1RC48BA1
|
NVIDIA B200 (1x B200-SXM-180GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1017 ROUGE2: 20.1125 ROUGEL: 29.9751 GEN_LEN: 4114386
| 19,739.8 | 2,467.475 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
d759ff5147d84bb9
|
closed
|
NVIDIA
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.0996 ROUGE2: 20.1107 ROUGEL: 29.9733 GEN_LEN: 4114082
| 19,233.8 | 2,404.225 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
1fc532fb2df64b4a
|
closed
|
NVIDIA
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1017 ROUGE2: 20.1125 ROUGEL: 29.9751 GEN_LEN: 4114386
| 19,739.8 | 2,467.475 |
Tokens/s
|
LLM
|
gptj-99.9
|
gptj-99.9
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
07c91a9edf4545be
|
closed
|
NVIDIA
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.0996 ROUGE2: 20.1107 ROUGEL: 29.9733 GEN_LEN: 4114082
| 19,233.8 | 2,404.225 |
Tokens/s
|
LLM
|
gptj-99.9
|
gptj-99.9
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
47dd2bf86fcf4d2a
|
closed
|
NVIDIA
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5437 ROUGE2: 22.1222 ROUGEL: 28.7396 TOKENS_PER_SAMPLE: 290.3
| 24,524.9 | 3,065.6125 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
7138c58328964e0e
|
closed
|
NVIDIA
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5435 ROUGE2: 22.122 ROUGEL: 28.7395 TOKENS_PER_SAMPLE: 290.3
| 21,605.8 | 2,700.725 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
8ffb0c3b5c7541c6
|
closed
|
NVIDIA
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5437 ROUGE2: 22.1222 ROUGEL: 28.7396 TOKENS_PER_SAMPLE: 290.3
| 24,524.9 | 3,065.6125 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
73728479d1ca4ea9
|
closed
|
NVIDIA
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5435 ROUGE2: 22.122 ROUGEL: 28.7395 TOKENS_PER_SAMPLE: 290.3
| 21,605.8 | 2,700.725 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
ab14d468cb284f0f
|
closed
|
NVIDIA
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 45.3448 ROUGE2: 23.2415 ROUGEL: 30.3087 TOKENS_PER_SAMPLE: 146.0 gsm8k_accuracy: 73.12 mbxp_accuracy: 59.88
| 52,415.9 | 6,551.9875 |
Tokens/s
|
LLM
|
mixtral-8x7b
|
mixtral-8x7b
| 47 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
5b547f884eba47f0
|
closed
|
NVIDIA
|
Offline
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 45.3537 ROUGE2: 23.2446 ROUGEL: 30.3114 TOKENS_PER_SAMPLE: 145.9 gsm8k_accuracy: 73.1 mbxp_accuracy: 59.88
| 50,795.6 | 6,349.45 |
Tokens/s
|
LLM
|
mixtral-8x7b
|
mixtral-8x7b
| 47 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
46bf49a0613049ce
|
closed
|
NVIDIA
|
Server
| 8 | 80 |
HBM3
|
NVIDIA H100-SXM-80GB
| 8 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA1
|
NVIDIA DGX H100 (8x H100-SXM-80GB, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1017 ROUGE2: 20.1125 ROUGEL: 29.9751 GEN_LEN: 4114386
| 2,627.69 | 2,627.69 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
available
|
410c8d452aef4024
|
closed
|
NVIDIA
|
Offline
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
NVIDIA GH200 NVL2 Platform (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1002 ROUGE2: 20.1114 ROUGEL: 29.9742 GEN_LEN: 4114259
| 2,513.38 | 2,513.38 |
Tokens/s
|
LLM
|
gptj-99
|
gptj-99
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
available
|
8945865c53b04a8d
|
closed
|
NVIDIA
|
Server
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
NVIDIA GH200 NVL2 Platform (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1017 ROUGE2: 20.1125 ROUGEL: 29.9751 GEN_LEN: 4114386
| 2,627.69 | 2,627.69 |
Tokens/s
|
LLM
|
gptj-99.9
|
gptj-99.9
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
available
|
fc1ba15a851041fe
|
closed
|
NVIDIA
|
Offline
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
NVIDIA GH200 NVL2 Platform (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 43.1002 ROUGE2: 20.1114 ROUGEL: 29.9742 GEN_LEN: 4114259
| 2,513.38 | 2,513.38 |
Tokens/s
|
LLM
|
gptj-99.9
|
gptj-99.9
| 6 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
available
|
6dc9caecd4c94263
|
closed
|
NVIDIA
|
Server
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
NVIDIA GH200 NVL2 Platform (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5631 ROUGE2: 22.1578 ROUGEL: 28.774 TOKENS_PER_SAMPLE: 291.2
| 4,067.52 | 4,067.52 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
available
|
7148ce73d794495b
|
closed
|
NVIDIA
|
Offline
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
NVIDIA GH200 NVL2 Platform (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.563 ROUGE2: 22.1576 ROUGEL: 28.7742 TOKENS_PER_SAMPLE: 291.3
| 3,883.67 | 3,883.67 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
available
|
027f1af4fd7d487f
|
closed
|
NVIDIA
|
Server
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
NVIDIA GH200 NVL2 Platform (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5631 ROUGE2: 22.1578 ROUGEL: 28.774 TOKENS_PER_SAMPLE: 291.2
| 4,067.52 | 4,067.52 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
available
|
738dc0bd6ae94f73
|
closed
|
NVIDIA
|
Offline
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
NVIDIA GH200 NVL2 Platform (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.563 ROUGE2: 22.1576 ROUGEL: 28.7742 TOKENS_PER_SAMPLE: 291.3
| 3,883.67 | 3,883.67 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
available
|
79bc025364d340b0
|
closed
|
NVIDIA
|
Server
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
NVIDIA GH200 NVL2 Platform (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 45.3572 ROUGE2: 23.2487 ROUGEL: 30.3161 TOKENS_PER_SAMPLE: 145.9 gsm8k_accuracy: 73.1 mbxp_accuracy: 59.92
| 8,021.46 | 8,021.46 |
Tokens/s
|
LLM
|
mixtral-8x7b
|
mixtral-8x7b
| 47 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
available
|
31a6a0c198954263
|
closed
|
NVIDIA
|
Offline
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
NVIDIA GH200 NVL2 Platform (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 45.3563 ROUGE2: 23.2486 ROUGEL: 30.3161 TOKENS_PER_SAMPLE: 145.9 gsm8k_accuracy: 73.1 mbxp_accuracy: 59.94
| 7,450.27 | 7,450.27 |
Tokens/s
|
LLM
|
mixtral-8x7b
|
mixtral-8x7b
| 47 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.2
|
available
|
fca13a038b7b44d3
|
closed
|
NVIDIA
|
Server
| 1 | 144 |
HBM3e
|
NVIDIA GH200 Grace Hopper Superchip 144GB
| 1 |
NVIDIA
| null | 72 | 1 | null |
NVIDIA Grace CPU
| null |
NVIDIA
|
InfiniBand
|
NVLink
| 480 |
15x 16DP (32GB) LPDDR5x
|
NVIDIA GH200 NVL2 Platform (1x GH200-144GB_aarch64, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5631 ROUGE2: 22.1578 ROUGEL: 28.774 TOKENS_PER_SAMPLE: 291.2
| 4,487.88 | 4,487.88 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
693be369354f42ea
|
closed
|
NVIDIA
|
Offline
| 1 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB-CTS
| 1 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA12
|
NVIDIA H200 (1x H200-SXM-141GB-CTS, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5629 ROUGE2: 22.1577 ROUGEL: 28.774 TOKENS_PER_SAMPLE: 291.3
| 4,202.3 | 4,202.3 |
Tokens/s
|
LLM
|
llama2-70b-99
|
llama2-70b-99
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
fa44b541511d4470
|
closed
|
NVIDIA
|
Server
| 1 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB-CTS
| 1 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA12
|
NVIDIA H200 (1x H200-SXM-141GB-CTS, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5631 ROUGE2: 22.1578 ROUGEL: 28.774 TOKENS_PER_SAMPLE: 291.2
| 4,487.88 | 4,487.88 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
f88fa3026f3540e3
|
closed
|
NVIDIA
|
Offline
| 1 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB-CTS
| 1 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA12
|
NVIDIA H200 (1x H200-SXM-141GB-CTS, TensorRT)
| 1 |
datacenter
|
mlperf-inference
|
4.1
| null |
ROUGE1: 44.5629 ROUGE2: 22.1577 ROUGEL: 28.774 TOKENS_PER_SAMPLE: 291.3
| 4,202.3 | 4,202.3 |
Tokens/s
|
LLM
|
llama2-70b-99.9
|
llama2-70b-99.9
| 70 |
fp16
|
TensorRT 10.2.0, CUDA 12.4
| null |
12.4
| null | null | null | null | null |
10.2.0
| null |
Ubuntu 22.04.4
|
available
|
e50c8f96e84c44a7
|
closed
|
NVIDIA
|
Server
| 1 | 141 |
HBM3e
|
NVIDIA H200-SXM-141GB-CTS
| 1 |
NVIDIA
| null | 56 | 2 | null |
Intel(R) Xeon(R) Platinum 8480C
| null |
Intel
|
NVLink-4
|
PCIe-5
| 2,048 |
32x 64GB MTC40F2046S1RC48BA12
|
NVIDIA H200 (1x H200-SXM-141GB-CTS, TensorRT)
| 1 |
datacenter
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.