Geekbench AI Score

4939
Single Precision Score
44889
Half Precision Score
59495
Quantized Score
Geekbench AI 1.7.0 for macOS AArch64

Result Information

User discobay
Upload Date March 17 2026 12:21 PM
Views 4

System Information

AI Information
Framework Core ML
Backend Neural Engine
Device Apple M5 Pro
System Information
Operating System macOS 26.3.1 (Build 25D2128)
Model Mac17,9
Model IDMac17,9
Motherboard Mac17,9
CPU Information
Name Apple M5 Pro
Topology 1 Processor, 18 Cores
Identifier Apple M5 Pro
Base Frequency 4.59 GHz
Cluster 16 Cores
Cluster 212 Cores
Memory Information
Size 64.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 2292
426.3 IPS
 
Image Classification (HP)
100% 21363
3.97 KIPS
 
Image Classification (Q)
97% 24964
4.66 KIPS
 
Image Segmentation (SP)
100% 2505
40.6 IPS
 
Image Segmentation (HP)
100% 26255
425.6 IPS
 
Image Segmentation (Q)
99% 36955
599.1 IPS
 
Pose Estimation (SP)
100% 10430
12.2 IPS
 
Pose Estimation (HP)
100% 180910
211.1 IPS
 
Pose Estimation (Q)
95% 333508
390.9 IPS
 
Object Detection (SP)
100% 2758
218.8 IPS
 
Object Detection (HP)
100% 24099
1.91 KIPS
 
Object Detection (Q)
91% 28573
2.28 KIPS
 
Face Detection (SP)
100% 5018
59.6 IPS
 
Face Detection (HP)
100% 49241
585.1 IPS
 
Face Detection (Q)
100% 69313
823.6 IPS
 
Depth Estimation (SP)
100% 9283
71.5 IPS
 
Depth Estimation (HP)
99% 136419
1.05 KIPS
 
Depth Estimation (Q)
85% 215317
1.68 KIPS
 
Style Transfer (SP)
100% 26747
34.4 IPS
 
Style Transfer (HP)
100% 285179
366.6 IPS
 
Style Transfer (Q)
96% 395226
510.0 IPS
 
Image Super-Resolution (SP)
100% 5983
220.9 IPS
 
Image Super-Resolution (HP)
100% 65850
2.43 KIPS
 
Image Super-Resolution (Q)
99% 101971
3.77 KIPS
 
Text Classification (SP)
100% 2918
3.90 KIPS
 
Text Classification (HP)
99% 5492
7.33 KIPS
 
Text Classification (Q)
93% 5411
7.26 KIPS
 
Machine Translation (SP)
100% 2407
41.5 IPS
 
Machine Translation (HP)
100% 19613
337.8 IPS
 
Machine Translation (Q)
98% 19423
335.7 IPS
 

Compare

Set Baseline Geekbench 6 now available