Geekbench AI Score

923
Single Precision Score
533
Half Precision Score
741
Quantized Score
Geekbench AI 1.1.0 for Windows x86 (64-bit)

Result Information

Upload Date September 16 2024 01:50 PM
Views 3

System Information

AI Information
Framework ONNX
Backend CPU
Device Intel Core i7-3770
System Information
Operating System Microsoft Windows 10 Pro (64-bit)
Model Acidanthera MacPro6,1
Motherboard Acidanthera Mac-F60DEB81FF30ACF6
Power PlanHigh performance
CPU Information
Name Intel Core i7-3770
Topology 1 Processor, 4 Cores, 8 Threads
Identifier GenuineIntel Family 6 Model 58 Stepping 9
Base Frequency 3.90 GHz
Cluster 14 Cores
Memory Information
Size 32.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 530
98.5 IPS
 
Image Classification (HP)
100% 214
39.8 IPS
 
Image Classification (Q)
96% 580
108.3 IPS
 
Image Segmentation (SP)
100% 490
7.94 IPS
 
Image Segmentation (HP)
100% 311
5.04 IPS
 
Image Segmentation (Q)
97% 538
8.74 IPS
 
Pose Estimation (SP)
100% 1444
1.69 IPS
 
Pose Estimation (HP)
100% 1106
1.29 IPS
 
Pose Estimation (Q)
87% 1334
1.58 IPS
 
Object Detection (SP)
100% 573
45.4 IPS
 
Object Detection (HP)
100% 275
21.8 IPS
 
Object Detection (Q)
90% 494
39.5 IPS
 
Face Detection (SP)
100% 1003
11.9 IPS
 
Face Detection (HP)
100% 615
7.31 IPS
 
Face Detection (Q)
96% 1106
13.2 IPS
 
Depth Estimation (SP)
100% 1665
12.8 IPS
 
Depth Estimation (HP)
99% 804
6.19 IPS
 
Depth Estimation (Q)
78% 1072
8.56 IPS
 
Style Transfer (SP)
100% 3454
4.44 IPS
 
Style Transfer (HP)
100% 2802
3.60 IPS
 
Style Transfer (Q)
87% 2949
3.84 IPS
 
Image Super-Resolution (SP)
100% 722
26.7 IPS
 
Image Super-Resolution (HP)
100% 613
22.6 IPS
 
Image Super-Resolution (Q)
97% 670
24.8 IPS
 
Text Classification (SP)
100% 667
890.4 IPS
 
Text Classification (HP)
100% 331
441.6 IPS
 
Text Classification (Q)
97% 259
346.8 IPS
 
Machine Translation (SP)
100% 757
13.0 IPS
 
Machine Translation (HP)
100% 329
5.66 IPS
 
Machine Translation (Q)
60% 405
9.03 IPS
 

Compare

Set Baseline Geekbench 6 now available