Geekbench AI Score

13009
Single Precision Score
14499
Half Precision Score
13360
Quantized Score
Geekbench AI 1.5.0 for macOS AArch64

Result Information

User yazkimkim
Upload Date September 08 2025 10:14 AM
Views 3

System Information

AI Information
Framework Core ML
Backend GPU
Device Apple M1 Ultra
System Information
Operating System macOS 15.6.1 (Build 24G90)
Model Mac Studio
Model IDMac13,2
Motherboard Mac13,2
CPU Information
Name Apple M1 Ultra
Topology 1 Processor, 20 Cores
Identifier Apple M1 Ultra
Base Frequency 3.22 GHz
Cluster 116 Cores
Cluster 24 Cores
Memory Information
Size 128.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 4873
906.2 IPS
 
Image Classification (HP)
100% 4741
881.6 IPS
 
Image Classification (Q)
96% 4351
812.4 IPS
 
Image Segmentation (SP)
100% 14177
229.8 IPS
 
Image Segmentation (HP)
100% 14599
236.7 IPS
 
Image Segmentation (Q)
99% 13713
222.3 IPS
 
Pose Estimation (SP)
100% 82037
95.7 IPS
 
Pose Estimation (HP)
100% 88308
103.0 IPS
 
Pose Estimation (Q)
96% 84275
98.7 IPS
 
Object Detection (SP)
100% 5034
399.3 IPS
 
Object Detection (HP)
100% 5129
406.9 IPS
 
Object Detection (Q)
92% 4644
370.7 IPS
 
Face Detection (SP)
100% 22385
266.0 IPS
 
Face Detection (HP)
100% 24885
295.7 IPS
 
Face Detection (Q)
100% 21718
258.1 IPS
 
Depth Estimation (SP)
100% 27370
210.9 IPS
 
Depth Estimation (HP)
99% 28208
217.3 IPS
 
Depth Estimation (Q)
86% 25320
197.8 IPS
 
Style Transfer (SP)
100% 133040
171.0 IPS
 
Style Transfer (HP)
100% 149742
192.5 IPS
 
Style Transfer (Q)
96% 149510
192.9 IPS
 
Image Super-Resolution (SP)
100% 15518
573.0 IPS
 
Image Super-Resolution (HP)
100% 17903
661.1 IPS
 
Image Super-Resolution (Q)
99% 14506
535.6 IPS
 
Text Classification (SP)
100% 2175
2.90 KIPS
 
Text Classification (HP)
100% 3756
5.01 KIPS
 
Text Classification (Q)
93% 3649
4.90 KIPS
 
Machine Translation (SP)
100% 1769
30.5 IPS
 
Machine Translation (HP)
100% 1854
31.9 IPS
 
Machine Translation (Q)
98% 1783
30.8 IPS
 

Compare

Set Baseline Geekbench 6 now available