Geekbench AI Score

10942
Single Precision Score
12385
Half Precision Score
11393
Quantized Score
Geekbench AI 1.5.0 for macOS AArch64

Result Information

User 1802
Upload Date September 15 2025 08:38 AM
Views 5

System Information

AI Information
Framework Core ML
Backend GPU
Device Apple M3 Pro
System Information
Operating System macOS 15.6.1 (Build 24G90)
Model MacBook Pro (14-inch, Nov 2023)
Model IDMac15,6
Motherboard Mac15,6
CPU Information
Name Apple M3 Pro
Topology 1 Processor, 11 Cores
Identifier Apple M3 Pro
Base Frequency 4.05 GHz
Cluster 15 Cores
Cluster 26 Cores
Memory Information
Size 18.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 5687
1.06 KIPS
 
Image Classification (HP)
100% 5938
1.10 KIPS
 
Image Classification (Q)
96% 5428
1.01 KIPS
 
Image Segmentation (SP)
100% 9227
149.6 IPS
 
Image Segmentation (HP)
100% 9872
160.0 IPS
 
Image Segmentation (Q)
99% 9771
158.4 IPS
 
Pose Estimation (SP)
100% 34922
40.7 IPS
 
Pose Estimation (HP)
100% 36046
42.1 IPS
 
Pose Estimation (Q)
96% 35685
41.8 IPS
 
Object Detection (SP)
100% 5536
439.1 IPS
 
Object Detection (HP)
100% 5972
473.7 IPS
 
Object Detection (Q)
92% 5380
429.4 IPS
 
Face Detection (SP)
100% 16004
190.2 IPS
 
Face Detection (HP)
100% 20067
238.4 IPS
 
Face Detection (Q)
100% 16234
192.9 IPS
 
Depth Estimation (SP)
100% 23297
179.5 IPS
 
Depth Estimation (HP)
99% 24470
188.5 IPS
 
Depth Estimation (Q)
86% 22728
177.5 IPS
 
Style Transfer (SP)
100% 77409
99.5 IPS
 
Style Transfer (HP)
100% 85323
109.7 IPS
 
Style Transfer (Q)
96% 83054
107.2 IPS
 
Image Super-Resolution (SP)
100% 13738
507.3 IPS
 
Image Super-Resolution (HP)
100% 14615
539.7 IPS
 
Image Super-Resolution (Q)
99% 12662
467.5 IPS
 
Text Classification (SP)
100% 2843
3.80 KIPS
 
Text Classification (HP)
100% 5079
6.78 KIPS
 
Text Classification (Q)
93% 5217
7.00 KIPS
 
Machine Translation (SP)
100% 2153
37.1 IPS
 
Machine Translation (HP)
100% 2165
37.3 IPS
 
Machine Translation (Q)
98% 1788
30.9 IPS
 

Compare

Set Baseline Geekbench 6 now available