Geekbench AI Score

2304
Single Precision Score
14965
Half Precision Score
16064
Quantized Score
Geekbench AI 1.0.0 for macOS AArch64

Result Information

User DragonLand
Upload Date August 15 2024 06:52 PM
Views 19

System Information

AI Information
Framework Core ML
Backend Neural Engine
Device Apple M2 Max
System Information
Operating System macOS 14.5 (Build 23F79)
Model MacBook Pro (14-inch, 2023)
Model IDMac14,5
Motherboard Mac14,5
CPU Information
Name Apple M2 Max
Topology 1 Processor, 12 Cores
Identifier Apple M2 Max
Base Frequency 3.43 GHz
Cluster 18 Cores
Cluster 24 Cores
Memory Information
Size 32.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 1118
207.9 IPS
 
Image Classification (HP)
100% 7964
1.48 KIPS
 
Image Classification (Q)
99% 8445
1.58 KIPS
 
Image Segmentation (SP)
100% 726
11.8 IPS
 
Image Segmentation (HP)
100% 5108
82.8 IPS
 
Image Segmentation (Q)
100% 5127
83.1 IPS
 
Pose Estimation (SP)
100% 11298
13.2 IPS
 
Pose Estimation (HP)
100% 92917
108.4 IPS
 
Pose Estimation (Q)
94% 88910
104.3 IPS
 
Object Detection (SP)
100% 1233
97.8 IPS
 
Object Detection (HP)
99% 9499
755.7 IPS
 
Object Detection (Q)
93% 10249
817.8 IPS
 
Face Detection (SP)
100% 2008
23.9 IPS
 
Face Detection (HP)
100% 11959
142.1 IPS
 
Face Detection (Q)
100% 13912
165.3 IPS
 
Depth Estimation (SP)
100% 3712
28.6 IPS
 
Depth Estimation (HP)
99% 57082
439.8 IPS
 
Depth Estimation (Q)
97% 61415
474.9 IPS
 
Style Transfer (SP)
100% 13293
17.1 IPS
 
Style Transfer (HP)
100% 147757
189.9 IPS
 
Style Transfer (Q)
100% 147170
189.2 IPS
 
Image Super-Resolution (SP)
100% 2273
83.9 IPS
 
Image Super-Resolution (HP)
100% 20991
775.1 IPS
 
Image Super-Resolution (Q)
98% 28410
1.05 KIPS
 
Text Classification (SP)
100% 883
1.18 KIPS
 
Text Classification (HP)
100% 3743
5.00 KIPS
 
Text Classification (Q)
93% 4112
5.52 KIPS
 
Machine Translation (SP)
100% 1878
32.3 IPS
 
Machine Translation (HP)
100% 1980
34.1 IPS
 
Machine Translation (Q)
98% 1975
34.1 IPS
 

Compare

Set Baseline Geekbench 6 now available