Geekbench AI Score

23125
Single Precision Score
33568
Half Precision Score
32498
Quantized Score
Geekbench AI 1.7.0 for macOS AArch64

Result Information

User blueweb
Upload Date October 28 2026 12:09 AM
Views 7

System Information

AI Information
Framework Core ML
Backend GPU
Device Apple M5 Max
System Information
Operating System macOS 27.0.1 (Build 26A434)
Model Mac17,14
Model IDMac17,14
Motherboard Mac17,14
CPU Information
Name Apple M5 Max
Topology 1 Processor, 18 Cores
Identifier Apple M5 Max
Base Frequency 4.60 GHz
Cluster 16 Cores
Cluster 212 Cores
Memory Information
Size 36.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 6201
1.15 KIPS
 
Image Classification (HP)
100% 6767
1.26 KIPS
 
Image Classification (Q)
97% 6717
1.25 KIPS
 
Image Segmentation (SP)
100% 24455
396.4 IPS
 
Image Segmentation (HP)
100% 33032
535.5 IPS
 
Image Segmentation (Q)
99% 32179
521.7 IPS
 
Pose Estimation (SP)
100% 97342
113.6 IPS
 
Pose Estimation (HP)
100% 227932
266.0 IPS
 
Pose Estimation (Q)
96% 218956
256.6 IPS
 
Object Detection (SP)
100% 8633
684.8 IPS
 
Object Detection (HP)
100% 10202
809.2 IPS
 
Object Detection (Q)
92% 10374
827.9 IPS
 
Face Detection (SP)
100% 38811
461.2 IPS
 
Face Detection (HP)
100% 52661
625.7 IPS
 
Face Detection (Q)
100% 48026
570.6 IPS
 
Depth Estimation (SP)
100% 47580
366.6 IPS
 
Depth Estimation (HP)
99% 59878
461.3 IPS
 
Depth Estimation (Q)
85% 58623
458.1 IPS
 
Style Transfer (SP)
100% 206036
264.9 IPS
 
Style Transfer (HP)
100% 392354
504.4 IPS
 
Style Transfer (Q)
96% 438121
565.4 IPS
 
Image Super-Resolution (SP)
100% 21141
780.6 IPS
 
Image Super-Resolution (HP)
100% 28493
1.05 KIPS
 
Image Super-Resolution (Q)
99% 21323
787.4 IPS
 
Text Classification (SP)
100% 6424
8.57 KIPS
 
Text Classification (HP)
99% 10259
13.7 KIPS
 
Text Classification (Q)
93% 10456
14.0 KIPS
 
Machine Translation (SP)
100% 6643
114.4 IPS
 
Machine Translation (HP)
100% 9664
166.5 IPS
 
Machine Translation (Q)
98% 9733
168.2 IPS
 

Compare

Set Baseline Geekbench AI now available