Geekbench ML Score

1710
ONNX DirectML Inference Score
Geekbench ML 0.6.0 for Windows AVX2

Result Information

User wxfisch
Upload Date August 15 2024 05:40 PM
Views 1

System Information

System Information
Operating System Microsoft Windows 11 Home (64-bit)
Model Gigabyte Technology Co., Ltd. X670 AORUS ELITE AX
Motherboard Gigabyte Technology Co., Ltd. X670 AORUS ELITE AX
Power PlanHigh performance
CPU Information
Name AMD Ryzen 5 7600X
Topology 1 Processor, 6 Cores, 12 Threads
Identifier AuthenticAMD Family 25 Model 97 Stepping 2
Base Frequency 4.70 GHz
Cluster 16 Cores
L1 Instruction Cache32.0 KB x 6
L1 Data Cache32.0 KB x 6
L2 Cache1.00 MB x 6
L3 Cache32.0 MB x 1
Memory Information
Size 32.00 GB
Inference Information
Framework ONNX
Backend DirectML
Device AMD Radeon(TM) Graphics

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 604
113.0 IPS
 
Image Classification (F16)
100% 603
112.8 IPS
 
Image Classification (I8)
100% 569
106.4 IPS
 
Image Segmentation (F32)
100% 1106
18.5 IPS
 
Image Segmentation (F16)
100% 1109
18.5 IPS
 
Image Segmentation (I8)
98% 961
16.1 IPS
 
Pose Estimation (F32)
100% 6396
7.75 IPS
 
Pose Estimation (F16)
100% 6409
7.76 IPS
 
Pose Estimation (I8)
100% 5744
6.96 IPS
 
Object Detection (F32)
100% 655
48.9 IPS
 
Object Detection (F16)
100% 656
49.0 IPS
 
Object Detection (I8)
62% 617
46.1 IPS
 
Face Detection (F32)
100% 1767
21.0 IPS
 
Face Detection (F16)
100% 1767
21.0 IPS
 
Face Detection (I8)
89% 1531
18.2 IPS
 
Depth Estimation (F32)
100% 3485
27.0 IPS
 
Depth Estimation (F16)
100% 3492
27.1 IPS
 
Depth Estimation (I8)
94% 3170
24.6 IPS
 
Style Transfer (F32)
100% 10974
14.4 IPS
 
Style Transfer (F16)
100% 10968
14.4 IPS
 
Style Transfer (I8)
98% 9723
12.8 IPS
 
Image Super-Resolution (F32)
100% 2789
99.6 IPS
 
Image Super-Resolution (F16)
100% 2795
99.8 IPS
 
Image Super-Resolution (I8)
99% 2118
75.6 IPS
 
Text Classification (F32)
100% 672
966.1 IPS
 
Text Classification (F16)
100% 662
951.3 IPS
 
Text Classification (I8)
98% 469
674.7 IPS
 
Machine Translation (F32)
100% 1162
21.4 IPS
 
Machine Translation (F16)
100% 1163
21.4 IPS
 
Machine Translation (I8)
70% 648
11.9 IPS
 

Compare

Set Baseline Geekbench 6 now available