Geekbench ML Score

759
ONNX DirectML Inference Score
Geekbench ML 0.6.0 for Windows AVX2

Result Information

User inc7007
Upload Date January 06 2024 05:48 PM
Views 1

System Information

System Information
Operating System Microsoft Windows 11 Pro (64-bit)
Model Microsoft Corporation Surface Book 2
Motherboard Microsoft Corporation Surface Book 2
Power PlanBalanced
CPU Information
Name Intel Core i5-7300U
Topology 1 Processor, 2 Cores, 4 Threads
Identifier GenuineIntel Family 6 Model 142 Stepping 9
Base Frequency 2.71 GHz
Cluster 12 Cores
L1 Instruction Cache32.0 KB x 2
L1 Data Cache32.0 KB x 2
L2 Cache256 KB x 2
L3 Cache3.00 MB x 1
Memory Information
Size 8.00 GB
Inference Information
Framework ONNX
Backend DirectML
Device Intel(R) HD Graphics 620

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 369
69.1 IPS
 
Image Classification (F16)
100% 362
67.8 IPS
 
Image Classification (I8)
100% 327
61.1 IPS
 
Image Segmentation (F32)
100% 457
7.63 IPS
 
Image Segmentation (F16)
100% 465
7.77 IPS
 
Image Segmentation (I8)
98% 422
7.04 IPS
 
Pose Estimation (F32)
100% 2183
2.64 IPS
 
Pose Estimation (F16)
100% 2182
2.64 IPS
 
Pose Estimation (I8)
100% 1977
2.39 IPS
 
Object Detection (F32)
100% 373
27.9 IPS
 
Object Detection (F16)
100% 361
26.9 IPS
 
Object Detection (I8)
54% 327
24.4 IPS
 
Face Detection (F32)
100% 708
8.42 IPS
 
Face Detection (F16)
100% 715
8.51 IPS
 
Face Detection (I8)
89% 573
6.82 IPS
 
Depth Estimation (F32)
100% 1571
12.2 IPS
 
Depth Estimation (F16)
100% 1550
12.0 IPS
 
Depth Estimation (I8)
95% 1388
10.8 IPS
 
Style Transfer (F32)
100% 5513
7.25 IPS
 
Style Transfer (F16)
100% 5546
7.29 IPS
 
Style Transfer (I8)
98% 4649
6.12 IPS
 
Image Super-Resolution (F32)
100% 1103
39.4 IPS
 
Image Super-Resolution (F16)
100% 1128
40.3 IPS
 
Image Super-Resolution (I8)
99% 857
30.6 IPS
 
Text Classification (F32)
100% 324
465.7 IPS
 
Text Classification (F16)
100% 375
538.3 IPS
 
Text Classification (I8)
98% 219
314.5 IPS
 
Machine Translation (F32)
100% 390
7.18 IPS
 
Machine Translation (F16)
100% 398
7.32 IPS
 
Machine Translation (I8)
70% 260
4.79 IPS
 

Compare

Set Baseline Geekbench 6 now available