Geekbench AI Score

4182
Single Precision Score
26716
Half Precision Score
29745
Quantized Score
Geekbench AI 1.2.0 for macOS AArch64

Result Information

User tim2o
Upload Date January 07 2025 12:35 AM
Views 4

System Information

AI Information
Framework Core ML
Backend Neural Engine
Device Apple M2 Max
System Information
Operating System macOS 15.2 (Build 24C101)
Model Mac Studio (2023)
Model IDMac14,13
Motherboard Mac14,13
CPU Information
Name Apple M2 Max
Topology 1 Processor, 12 Cores
Identifier Apple M2 Max
Base Frequency 3.48 GHz
Cluster 18 Cores
Cluster 24 Cores
Memory Information
Size 32.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 1410
262.1 IPS
 
Image Classification (HP)
100% 13974
2.60 KIPS
 
Image Classification (Q)
96% 17168
3.21 KIPS
 
Image Segmentation (SP)
100% 1673
27.1 IPS
 
Image Segmentation (HP)
100% 19517
316.4 IPS
 
Image Segmentation (Q)
99% 24712
400.6 IPS
 
Pose Estimation (SP)
100% 13155
15.3 IPS
 
Pose Estimation (HP)
100% 106323
124.1 IPS
 
Pose Estimation (Q)
95% 104489
122.5 IPS
 
Object Detection (SP)
100% 1699
134.8 IPS
 
Object Detection (HP)
99% 15425
1.23 KIPS
 
Object Detection (Q)
91% 17019
1.36 KIPS
 
Face Detection (SP)
100% 3662
43.5 IPS
 
Face Detection (HP)
100% 34941
415.2 IPS
 
Face Detection (Q)
100% 41868
497.5 IPS
 
Depth Estimation (SP)
100% 6434
49.6 IPS
 
Depth Estimation (HP)
99% 87285
672.5 IPS
 
Depth Estimation (Q)
85% 108669
849.2 IPS
 
Style Transfer (SP)
100% 24909
32.0 IPS
 
Style Transfer (HP)
100% 203201
261.2 IPS
 
Style Transfer (Q)
96% 191941
247.7 IPS
 
Image Super-Resolution (SP)
100% 5106
188.5 IPS
 
Image Super-Resolution (HP)
100% 41236
1.52 KIPS
 
Image Super-Resolution (Q)
99% 48722
1.80 KIPS
 
Text Classification (SP)
100% 2292
3.06 KIPS
 
Text Classification (HP)
100% 4051
5.41 KIPS
 
Text Classification (Q)
93% 3960
5.31 KIPS
 
Machine Translation (SP)
100% 4520
77.8 IPS
 
Machine Translation (HP)
100% 4001
68.9 IPS
 
Machine Translation (Q)
98% 4266
73.7 IPS
 

Compare

Set Baseline Geekbench 6 now available