Geekbench AI Score

4842
Single Precision Score
27416
Half Precision Score
30903
Quantized Score
Geekbench AI 1.0.0 for macOS AArch64

Result Information

User wp4
Upload Date November 21 2026 05:57 PM
Views 3

System Information

AI Information
Framework Core ML
Backend Neural Engine
Device Apple M3 Max
System Information
Operating System macOS 27.2.1 (Build 26B131)
Model MacBook Pro (16-inch, Nov 2023)
Model IDMac15,11
Motherboard Mac15,11
CPU Information
Name Apple M3 Max
Topology 1 Processor, 14 Cores
Identifier Apple M3 Max
Base Frequency 4.06 GHz
Cluster 110 Cores
Cluster 24 Cores
Memory Information
Size 36.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 1459
271.3 IPS
 
Image Classification (HP)
100% 14995
2.79 KIPS
 
Image Classification (Q)
99% 17182
3.20 KIPS
 
Image Segmentation (SP)
100% 1851
30.0 IPS
 
Image Segmentation (HP)
100% 16745
271.5 IPS
 
Image Segmentation (Q)
100% 16484
267.2 IPS
 
Pose Estimation (SP)
100% 14192
16.6 IPS
 
Pose Estimation (HP)
100% 114991
134.2 IPS
 
Pose Estimation (Q)
94% 113531
133.2 IPS
 
Object Detection (SP)
100% 1856
147.2 IPS
 
Object Detection (HP)
100% 15234
1.21 KIPS
 
Object Detection (Q)
89% 20235
1.62 KIPS
 
Face Detection (SP)
100% 4049
48.1 IPS
 
Face Detection (HP)
100% 25350
301.2 IPS
 
Face Detection (Q)
100% 40396
480.0 IPS
 
Depth Estimation (SP)
100% 7062
54.4 IPS
 
Depth Estimation (HP)
99% 82424
635.0 IPS
 
Depth Estimation (Q)
97% 85785
663.3 IPS
 
Style Transfer (SP)
100% 30333
39.0 IPS
 
Style Transfer (HP)
100% 172232
221.4 IPS
 
Style Transfer (Q)
100% 173430
222.9 IPS
 
Image Super-Resolution (SP)
100% 5516
203.7 IPS
 
Image Super-Resolution (HP)
100% 37932
1.40 KIPS
 
Image Super-Resolution (Q)
98% 45446
1.68 KIPS
 
Text Classification (SP)
100% 5268
7.03 KIPS
 
Text Classification (HP)
99% 8575
11.4 KIPS
 
Text Classification (Q)
93% 8818
11.8 KIPS
 
Machine Translation (SP)
100% 3956
68.1 IPS
 
Machine Translation (HP)
100% 4660
80.3 IPS
 
Machine Translation (Q)
98% 5070
87.6 IPS
 

Compare

Set Baseline Geekbench AI now available