Geekbench AI Score

15601
Single Precision Score
19726
Half Precision Score
17745
Quantized Score
Geekbench AI 1.1.0 for macOS AArch64

Result Information

Upload Date November 21 2024 06:56 AM
Views 2

System Information

AI Information
Framework Core ML
Backend GPU
Device Apple M3 Max
System Information
Operating System macOS 15.1.1 (Build 24B91)
Model MacBook Pro (14-inch, Nov 2023)
Model IDMac15,10
Motherboard Mac15,10
CPU Information
Name Apple M3 Max
Topology 1 Processor, 14 Cores
Identifier Apple M3 Max
Base Frequency 4.04 GHz
Cluster 110 Cores
Cluster 24 Cores
Memory Information
Size 36.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 7283
1.35 KIPS
 
Image Classification (HP)
100% 7726
1.44 KIPS
 
Image Classification (Q)
97% 6782
1.27 KIPS
 
Image Segmentation (SP)
100% 13775
223.3 IPS
 
Image Segmentation (HP)
100% 14623
237.1 IPS
 
Image Segmentation (Q)
99% 15164
245.8 IPS
 
Pose Estimation (SP)
100% 66635
77.8 IPS
 
Pose Estimation (HP)
100% 69259
80.8 IPS
 
Pose Estimation (Q)
95% 65813
77.1 IPS
 
Object Detection (SP)
100% 7017
556.6 IPS
 
Object Detection (HP)
100% 7557
599.4 IPS
 
Object Detection (Q)
91% 6375
509.4 IPS
 
Face Detection (SP)
100% 25867
307.4 IPS
 
Face Detection (HP)
100% 30434
361.6 IPS
 
Face Detection (Q)
100% 23451
278.7 IPS
 
Depth Estimation (SP)
100% 33152
255.4 IPS
 
Depth Estimation (HP)
99% 35717
275.2 IPS
 
Depth Estimation (Q)
85% 28780
224.9 IPS
 
Style Transfer (SP)
100% 143120
184.0 IPS
 
Style Transfer (HP)
100% 161515
207.6 IPS
 
Style Transfer (Q)
96% 149889
193.4 IPS
 
Image Super-Resolution (SP)
100% 21481
793.2 IPS
 
Image Super-Resolution (HP)
100% 22812
842.4 IPS
 
Image Super-Resolution (Q)
99% 18590
686.4 IPS
 
Text Classification (SP)
100% 2915
3.89 KIPS
 
Text Classification (HP)
100% 5252
7.01 KIPS
 
Text Classification (Q)
93% 5127
6.88 KIPS
 
Machine Translation (SP)
100% 2370
40.8 IPS
 
Machine Translation (HP)
98% 7175
124.0 IPS
 
Machine Translation (Q)
98% 7444
128.6 IPS
 

Compare

Set Baseline Geekbench 6 now available