Geekbench AI Score

4121
Single Precision Score
24327
Half Precision Score
27382
Quantized Score
Geekbench AI 1.0.0 for macOS AArch64

Result Information

User mkninc
Upload Date November 09 2024 06:57 PM
Views 2

System Information

AI Information
Framework Core ML
Backend Neural Engine
Device Apple M2 Pro
System Information
Operating System macOS 15.1 (Build 24B83)
Model Mac mini (2023)
Model IDMac14,12
Motherboard Mac14,12
CPU Information
Name Apple M2 Pro
Topology 1 Processor, 12 Cores
Identifier Apple M2 Pro
Base Frequency 3.45 GHz
Cluster 18 Cores
Cluster 24 Cores
Memory Information
Size 32.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 1371
255.0 IPS
 
Image Classification (HP)
100% 14037
2.61 KIPS
 
Image Classification (Q)
99% 16238
3.03 KIPS
 
Image Segmentation (SP)
100% 1694
27.5 IPS
 
Image Segmentation (HP)
100% 15465
250.7 IPS
 
Image Segmentation (Q)
100% 15670
254.0 IPS
 
Pose Estimation (SP)
100% 12606
14.7 IPS
 
Pose Estimation (HP)
100% 102656
119.8 IPS
 
Pose Estimation (Q)
94% 99923
117.2 IPS
 
Object Detection (SP)
100% 1674
132.8 IPS
 
Object Detection (HP)
99% 14062
1.12 KIPS
 
Object Detection (Q)
91% 17096
1.37 KIPS
 
Face Detection (SP)
100% 3531
42.0 IPS
 
Face Detection (HP)
100% 22993
273.2 IPS
 
Face Detection (Q)
100% 35100
417.1 IPS
 
Depth Estimation (SP)
100% 6211
47.8 IPS
 
Depth Estimation (HP)
99% 73696
567.8 IPS
 
Depth Estimation (Q)
97% 82422
637.3 IPS
 
Style Transfer (SP)
100% 24977
32.1 IPS
 
Style Transfer (HP)
100% 155559
200.0 IPS
 
Style Transfer (Q)
100% 157534
202.5 IPS
 
Image Super-Resolution (SP)
100% 5053
186.6 IPS
 
Image Super-Resolution (HP)
100% 42632
1.57 KIPS
 
Image Super-Resolution (Q)
98% 49783
1.84 KIPS
 
Text Classification (SP)
100% 2278
3.04 KIPS
 
Text Classification (HP)
100% 4030
5.38 KIPS
 
Text Classification (Q)
93% 3988
5.35 KIPS
 
Machine Translation (SP)
100% 4576
78.8 IPS
 
Machine Translation (HP)
100% 5117
88.1 IPS
 
Machine Translation (Q)
98% 6026
104.1 IPS
 

Compare

Set Baseline Geekbench 6 now available