Geekbench AI Score

3413
Single Precision Score
26698
Half Precision Score
29685
Quantized Score
Geekbench AI 1.5.0 for macOS AArch64

Result Information

User Serpadu
Upload Date September 11 2025 12:38 PM
Views 5

System Information

AI Information
Framework Core ML
Backend Neural Engine
Device Apple M2
System Information
Operating System macOS 15.6.1 (Build 24G90)
Model MacBook Air (15-inch, 2023)
Model IDMac14,15
Motherboard Mac14,15
CPU Information
Name Apple M2
Topology 1 Processor, 8 Cores
Identifier Apple M2
Base Frequency 3.49 GHz
Cluster 14 Cores
Cluster 24 Cores
Memory Information
Size 8.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 1366
254.1 IPS
 
Image Classification (HP)
100% 14664
2.73 KIPS
 
Image Classification (Q)
96% 15393
2.87 KIPS
 
Image Segmentation (SP)
100% 1419
23.0 IPS
 
Image Segmentation (HP)
100% 19640
318.4 IPS
 
Image Segmentation (Q)
99% 24104
390.7 IPS
 
Pose Estimation (SP)
100% 6813
7.95 IPS
 
Pose Estimation (HP)
100% 105631
123.3 IPS
 
Pose Estimation (Q)
95% 106504
124.8 IPS
 
Object Detection (SP)
100% 1513
120.0 IPS
 
Object Detection (HP)
99% 15614
1.24 KIPS
 
Object Detection (Q)
91% 17579
1.40 KIPS
 
Face Detection (SP)
100% 3041
36.1 IPS
 
Face Detection (HP)
100% 33249
395.1 IPS
 
Face Detection (Q)
100% 42106
500.3 IPS
 
Depth Estimation (SP)
100% 5518
42.5 IPS
 
Depth Estimation (HP)
99% 87555
674.6 IPS
 
Depth Estimation (Q)
85% 107991
843.9 IPS
 
Style Transfer (SP)
100% 18588
23.9 IPS
 
Style Transfer (HP)
100% 203827
262.0 IPS
 
Style Transfer (Q)
96% 194327
250.8 IPS
 
Image Super-Resolution (SP)
100% 3947
145.8 IPS
 
Image Super-Resolution (HP)
100% 39748
1.47 KIPS
 
Image Super-Resolution (Q)
99% 51359
1.90 KIPS
 
Text Classification (SP)
100% 2349
3.14 KIPS
 
Text Classification (HP)
100% 3822
5.10 KIPS
 
Text Classification (Q)
93% 3807
5.11 KIPS
 
Machine Translation (SP)
100% 3721
64.1 IPS
 
Machine Translation (HP)
100% 4298
74.0 IPS
 
Machine Translation (Q)
98% 4428
76.5 IPS
 

Compare

Set Baseline Geekbench 6 now available