Geekbench AI Score

11869
Single Precision Score
13770
Half Precision Score
12684
Quantized Score
Geekbench AI 1.5.0 for macOS AArch64

Result Information

User BrantB
Upload Date December 11 2025 04:49 AM
Views 13

System Information

AI Information
Framework Core ML
Backend GPU
Device Apple M3 Pro
System Information
Operating System macOS 26.1 (Build 25B78)
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 4.03 GHz
Cluster 15 Cores
Cluster 26 Cores
Memory Information
Size 18.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 5817
1.08 KIPS
 
Image Classification (HP)
100% 6274
1.17 KIPS
 
Image Classification (Q)
96% 5474
1.02 KIPS
 
Image Segmentation (SP)
100% 9359
151.7 IPS
 
Image Segmentation (HP)
100% 10123
164.1 IPS
 
Image Segmentation (Q)
99% 9631
156.1 IPS
 
Pose Estimation (SP)
100% 33750
39.4 IPS
 
Pose Estimation (HP)
100% 37768
44.1 IPS
 
Pose Estimation (Q)
96% 36589
42.9 IPS
 
Object Detection (SP)
100% 5561
441.1 IPS
 
Object Detection (HP)
100% 5888
467.1 IPS
 
Object Detection (Q)
92% 5456
435.5 IPS
 
Face Detection (SP)
100% 16599
197.2 IPS
 
Face Detection (HP)
100% 20599
244.8 IPS
 
Face Detection (Q)
100% 16789
199.5 IPS
 
Depth Estimation (SP)
100% 22886
176.3 IPS
 
Depth Estimation (HP)
99% 25321
195.1 IPS
 
Depth Estimation (Q)
86% 23058
180.1 IPS
 
Style Transfer (SP)
100% 76535
98.4 IPS
 
Style Transfer (HP)
100% 87520
112.5 IPS
 
Style Transfer (Q)
96% 85067
109.8 IPS
 
Image Super-Resolution (SP)
100% 13722
506.7 IPS
 
Image Super-Resolution (HP)
100% 15491
572.0 IPS
 
Image Super-Resolution (Q)
99% 12734
470.2 IPS
 
Text Classification (SP)
100% 2970
3.96 KIPS
 
Text Classification (HP)
99% 4987
6.66 KIPS
 
Text Classification (Q)
93% 4961
6.66 KIPS
 
Machine Translation (SP)
100% 4585
79.0 IPS
 
Machine Translation (HP)
100% 4923
84.8 IPS
 
Machine Translation (Q)
98% 4925
85.1 IPS
 

Compare

Set Baseline Geekbench 6 now available