Geekbench AI Score

22206
Single Precision Score
37625
Half Precision Score
37822
Quantized Score
Geekbench AI 1.7.0 for macOS AArch64

Result Information

User dean0x7d
Upload Date November 02 2026 04:46 PM
Views 8

System Information

AI Information
Framework Core ML
Backend GPU
Device Apple M5 Pro
System Information
Operating System macOS 27.2 (Build 26B123)
Model Mac17,9
Model IDMac17,9
Motherboard Mac17,9
CPU Information
Name Apple M5 Pro
Topology 1 Processor, 18 Cores
Identifier Apple M5 Pro
Base Frequency 4.60 GHz
Cluster 16 Cores
Cluster 212 Cores
Memory Information
Size 48.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 9122
1.70 KIPS
 
Image Classification (HP)
100% 12544
2.33 KIPS
 
Image Classification (Q)
99% 12131
2.26 KIPS
 
Image Segmentation (SP)
100% 19293
312.7 IPS
 
Image Segmentation (HP)
100% 32933
533.9 IPS
 
Image Segmentation (Q)
99% 31643
513.0 IPS
 
Pose Estimation (SP)
100% 70148
81.9 IPS
 
Pose Estimation (HP)
100% 185198
216.1 IPS
 
Pose Estimation (Q)
96% 244806
286.8 IPS
 
Object Detection (SP)
100% 10166
806.4 IPS
 
Object Detection (HP)
100% 13797
1.09 KIPS
 
Object Detection (Q)
92% 14113
1.13 KIPS
 
Face Detection (SP)
100% 33496
398.0 IPS
 
Face Detection (HP)
100% 54295
645.1 IPS
 
Face Detection (Q)
100% 50844
604.1 IPS
 
Depth Estimation (SP)
100% 40519
312.2 IPS
 
Depth Estimation (HP)
99% 62431
481.0 IPS
 
Depth Estimation (Q)
85% 65017
508.0 IPS
 
Style Transfer (SP)
100% 155945
200.5 IPS
 
Style Transfer (HP)
100% 367140
472.0 IPS
 
Style Transfer (Q)
96% 474096
611.8 IPS
 
Image Super-Resolution (SP)
100% 26175
966.5 IPS
 
Image Super-Resolution (HP)
100% 43847
1.62 KIPS
 
Image Super-Resolution (Q)
99% 28568
1.05 KIPS
 
Text Classification (SP)
100% 6428
8.58 KIPS
 
Text Classification (HP)
99% 10212
13.6 KIPS
 
Text Classification (Q)
93% 10400
14.0 KIPS
 
Machine Translation (SP)
100% 6525
112.4 IPS
 
Machine Translation (HP)
100% 9666
166.5 IPS
 
Machine Translation (Q)
98% 9701
167.7 IPS
 

Compare

Set Baseline Geekbench AI now available