Geekbench AI Score

6026
Single Precision Score
36826
Half Precision Score
49465
Quantized Score
Geekbench AI 1.3.0 for macOS AArch64

Result Information

User l3uzz
Upload Date October 31 2025 03:38 PM
Views 10

System Information

AI Information
Framework Core ML
Backend Neural Engine
Device Apple M4 Max
System Information
Operating System macOS 26.0.1 (Build 25A362)
Model Mac16,9
Model IDMac16,9
Motherboard Mac16,9
CPU Information
Name Apple M4 Max
Topology 1 Processor, 16 Cores
Identifier Apple M4 Max
Base Frequency 4.51 GHz
Cluster 112 Cores
Cluster 24 Cores
Memory Information
Size 128.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 2191
407.4 IPS
 
Image Classification (HP)
100% 17187
3.20 KIPS
 
Image Classification (Q)
97% 21075
3.93 KIPS
 
Image Segmentation (SP)
100% 2420
39.2 IPS
 
Image Segmentation (HP)
100% 22888
371.0 IPS
 
Image Segmentation (Q)
99% 31843
516.2 IPS
 
Pose Estimation (SP)
100% 18361
21.4 IPS
 
Pose Estimation (HP)
100% 122339
142.7 IPS
 
Pose Estimation (Q)
95% 231604
271.5 IPS
 
Object Detection (SP)
100% 2745
217.8 IPS
 
Object Detection (HP)
100% 19395
1.54 KIPS
 
Object Detection (Q)
91% 23549
1.88 KIPS
 
Face Detection (SP)
100% 5338
63.4 IPS
 
Face Detection (HP)
100% 44408
527.7 IPS
 
Face Detection (Q)
100% 62894
747.3 IPS
 
Depth Estimation (SP)
100% 9088
70.0 IPS
 
Depth Estimation (HP)
99% 108363
834.9 IPS
 
Depth Estimation (Q)
85% 175006
1.37 KIPS
 
Style Transfer (SP)
100% 31640
40.7 IPS
 
Style Transfer (HP)
100% 230977
296.9 IPS
 
Style Transfer (Q)
96% 318748
411.3 IPS
 
Image Super-Resolution (SP)
100% 7098
262.1 IPS
 
Image Super-Resolution (HP)
100% 48692
1.80 KIPS
 
Image Super-Resolution (Q)
99% 76196
2.81 KIPS
 
Text Classification (SP)
100% 3259
4.35 KIPS
 
Text Classification (HP)
99% 5625
7.51 KIPS
 
Text Classification (Q)
93% 5573
7.48 KIPS
 
Machine Translation (SP)
100% 6661
114.7 IPS
 
Machine Translation (HP)
100% 16145
278.1 IPS
 
Machine Translation (Q)
98% 16084
278.0 IPS
 

Compare

Set Baseline Geekbench 6 now available