Geekbench AI Score

23522
Single Precision Score
26223
Half Precision Score
24565
Quantized Score
Geekbench AI 1.5.0 for macOS AArch64

Result Information

User shafnitz
Upload Date November 08 2025 03:52 AM
Views 6

System Information

AI Information
Framework Core ML
Backend GPU
Device Apple M4 Max
System Information
Operating System macOS 15.7.2 (Build 24G325)
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 64.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 10519
1.96 KIPS
 
Image Classification (HP)
100% 10711
1.99 KIPS
 
Image Classification (Q)
96% 9719
1.82 KIPS
 
Image Segmentation (SP)
100% 23424
379.7 IPS
 
Image Segmentation (HP)
100% 24339
394.6 IPS
 
Image Segmentation (Q)
99% 23017
373.1 IPS
 
Pose Estimation (SP)
100% 106428
124.2 IPS
 
Pose Estimation (HP)
100% 109347
127.6 IPS
 
Pose Estimation (Q)
96% 105173
123.2 IPS
 
Object Detection (SP)
100% 8769
695.6 IPS
 
Object Detection (HP)
100% 10162
806.0 IPS
 
Object Detection (Q)
92% 9466
755.6 IPS
 
Face Detection (SP)
100% 39370
467.8 IPS
 
Face Detection (HP)
100% 42076
500.0 IPS
 
Face Detection (Q)
100% 36849
437.9 IPS
 
Depth Estimation (SP)
100% 44994
346.7 IPS
 
Depth Estimation (HP)
99% 47453
365.6 IPS
 
Depth Estimation (Q)
86% 43597
340.5 IPS
 
Style Transfer (SP)
100% 220824
283.9 IPS
 
Style Transfer (HP)
100% 233806
300.6 IPS
 
Style Transfer (Q)
96% 228680
295.1 IPS
 
Image Super-Resolution (SP)
100% 30689
1.13 KIPS
 
Image Super-Resolution (HP)
100% 31578
1.17 KIPS
 
Image Super-Resolution (Q)
99% 27475
1.01 KIPS
 
Text Classification (SP)
100% 2973
3.97 KIPS
 
Text Classification (HP)
100% 5467
7.30 KIPS
 
Text Classification (Q)
93% 5396
7.24 KIPS
 
Machine Translation (SP)
100% 6319
108.9 IPS
 
Machine Translation (HP)
100% 6587
113.5 IPS
 
Machine Translation (Q)
98% 6597
114.0 IPS
 

Compare

Set Baseline Geekbench 6 now available