Geekbench AI Score

4197
Single Precision Score
7403
Half Precision Score
5792
Quantized Score
Geekbench AI 1.7.0 Pro for macOS AArch64

Result Information

User wolfspyre
Upload Date October 15 2026 11:44 PM
Views 9
Notes Post OS Update + metal install

System Information

AI Information
Framework Core ML
Backend CPU
Device Apple M2 Max
System Information
Operating System macOS 27.2 (Build 26B5112a)
Model MacBook Pro (16-inch, 2023)
Model IDMac14,6
Motherboard Mac14,6
CPU Information
Name Apple M2 Max
Topology 1 Processor, 12 Cores
Identifier Apple M2 Max
Base Frequency 3.67 GHz
Cluster 18 Cores
Cluster 24 Cores
Memory Information
Size 96.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 1345
250.2 IPS
 
Image Classification (HP)
100% 2361
439.0 IPS
 
Image Classification (Q)
96% 1697
317.0 IPS
 
Image Segmentation (SP)
100% 1714
27.8 IPS
 
Image Segmentation (HP)
100% 2594
42.1 IPS
 
Image Segmentation (Q)
99% 2942
47.7 IPS
 
Pose Estimation (SP)
100% 10452
12.2 IPS
 
Pose Estimation (HP)
100% 25071
29.3 IPS
 
Pose Estimation (Q)
95% 19699
23.1 IPS
 
Object Detection (SP)
100% 1619
128.5 IPS
 
Object Detection (HP)
97% 2805
223.3 IPS
 
Object Detection (Q)
94% 2011
160.3 IPS
 
Face Detection (SP)
100% 3595
42.7 IPS
 
Face Detection (HP)
100% 6205
73.7 IPS
 
Face Detection (Q)
100% 3458
41.1 IPS
 
Depth Estimation (SP)
100% 5838
45.0 IPS
 
Depth Estimation (HP)
98% 12081
93.4 IPS
 
Depth Estimation (Q)
85% 8191
64.0 IPS
 
Style Transfer (SP)
100% 24263
31.2 IPS
 
Style Transfer (HP)
100% 41910
53.9 IPS
 
Style Transfer (Q)
96% 32471
41.9 IPS
 
Image Super-Resolution (SP)
100% 4520
166.9 IPS
 
Image Super-Resolution (HP)
100% 6759
249.6 IPS
 
Image Super-Resolution (Q)
99% 4399
162.4 IPS
 
Text Classification (SP)
100% 4447
5.94 KIPS
 
Text Classification (HP)
99% 6863
9.16 KIPS
 
Text Classification (Q)
93% 6778
9.10 KIPS
 
Machine Translation (SP)
100% 4252
73.2 IPS
 
Machine Translation (HP)
98% 7879
136.1 IPS
 
Machine Translation (Q)
98% 7840
135.5 IPS
 

Compare

Set Baseline Geekbench AI now available