Geekbench ML Score

1771
ONNX DirectML Inference Score
Geekbench ML 0.6.0 for Windows AVX2

Result Information

User dynamiteCobra88
Upload Date November 21 2024 10:31 AM
Views 4

System Information

System Information
Operating System Microsoft Windows 11 Pro (64-bit)
Model Microsoft Corporation Surface Pro 10 for Business
Motherboard Microsoft Corporation Surface Pro 10 for Business
Power PlanBalanced
CPU Information
Name Intel(R) Core(TM) Ultra 5 135U
Topology 1 Processor, 12 Cores, 14 Threads
Identifier GenuineIntel Family 6 Model 170 Stepping 4
Base Frequency 2.10 GHz
Cluster 12 Cores
Cluster 210 Cores
L1 Instruction Cache64.0 KB x 7
L1 Data Cache48.0 KB x 7
L2 Cache2.00 MB x 1
L3 Cache12.0 MB x 1
Memory Information
Size 16.00 GB
Inference Information
Framework ONNX
Backend DirectML
Device Intel(R) Graphics

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 645
120.7 IPS
 
Image Classification (F16)
100% 644
120.6 IPS
 
Image Classification (I8)
100% 591
110.6 IPS
 
Image Segmentation (F32)
100% 791
13.2 IPS
 
Image Segmentation (F16)
100% 786
13.1 IPS
 
Image Segmentation (I8)
98% 708
11.8 IPS
 
Pose Estimation (F32)
100% 7026
8.51 IPS
 
Pose Estimation (F16)
100% 7069
8.56 IPS
 
Pose Estimation (I8)
100% 6301
7.63 IPS
 
Object Detection (F32)
100% 636
47.5 IPS
 
Object Detection (F16)
100% 638
47.6 IPS
 
Object Detection (I8)
54% 581
43.4 IPS
 
Face Detection (F32)
100% 1029
12.2 IPS
 
Face Detection (F16)
100% 1030
12.3 IPS
 
Face Detection (I8)
89% 919
10.9 IPS
 
Depth Estimation (F32)
100% 3618
28.1 IPS
 
Depth Estimation (F16)
100% 3614
28.0 IPS
 
Depth Estimation (I8)
95% 3150
24.4 IPS
 
Style Transfer (F32)
100% 19492
25.6 IPS
 
Style Transfer (F16)
100% 19552
25.7 IPS
 
Style Transfer (I8)
98% 16100
21.2 IPS
 
Image Super-Resolution (F32)
100% 3136
112.0 IPS
 
Image Super-Resolution (F16)
100% 3956
141.3 IPS
 
Image Super-Resolution (I8)
99% 2777
99.2 IPS
 
Text Classification (F32)
90% 636
914.6 IPS
 
Text Classification (F16)
90% 619
889.8 IPS
 
Text Classification (I8)
90% 443
637.3 IPS
 
Machine Translation (F32)
100% 1748
32.2 IPS
 
Machine Translation (F16)
100% 1740
32.0 IPS
 
Machine Translation (I8)
70% 791
14.6 IPS
 

Compare

Set Baseline Geekbench 6 now available