Geekbench AI Score

4413
Single Precision Score
37183
Half Precision Score
52958
Quantized Score
Geekbench AI 1.3.0 for macOS AArch64

Result Information

User CircuitSphere
Upload Date March 20 2025 12:49 AM
Views 21

System Information

AI Information
Framework Core ML
Backend Neural Engine
Device Apple M4
System Information
Operating System macOS 15.3.2 (Build 24D81)
Model Mac mini (2024)
Model IDMac16,10
Motherboard Mac16,10
CPU Information
Name Apple M4
Topology 1 Processor, 10 Cores
Identifier Apple M4
Base Frequency 4.25 GHz
Cluster 14 Cores
Cluster 26 Cores
Memory Information
Size 16.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 1871
347.9 IPS
 
Image Classification (HP)
100% 17828
3.32 KIPS
 
Image Classification (Q)
97% 26360
4.92 KIPS
 
Image Segmentation (SP)
100% 1971
31.9 IPS
 
Image Segmentation (HP)
100% 22329
362.0 IPS
 
Image Segmentation (Q)
99% 30705
497.8 IPS
 
Pose Estimation (SP)
100% 9299
10.9 IPS
 
Pose Estimation (HP)
100% 124998
145.9 IPS
 
Pose Estimation (Q)
95% 234993
275.4 IPS
 
Object Detection (SP)
100% 2153
170.8 IPS
 
Object Detection (HP)
100% 18968
1.50 KIPS
 
Object Detection (Q)
91% 26283
2.10 KIPS
 
Face Detection (SP)
100% 3598
42.8 IPS
 
Face Detection (HP)
100% 45869
545.0 IPS
 
Face Detection (Q)
100% 66461
789.7 IPS
 
Depth Estimation (SP)
100% 7830
60.3 IPS
 
Depth Estimation (HP)
99% 100221
772.1 IPS
 
Depth Estimation (Q)
85% 170527
1.33 KIPS
 
Style Transfer (SP)
100% 23382
30.1 IPS
 
Style Transfer (HP)
100% 240470
309.1 IPS
 
Style Transfer (Q)
96% 305000
393.6 IPS
 
Image Super-Resolution (SP)
100% 5342
197.3 IPS
 
Image Super-Resolution (HP)
100% 49261
1.82 KIPS
 
Image Super-Resolution (Q)
99% 87974
3.25 KIPS
 
Text Classification (SP)
100% 2933
3.91 KIPS
 
Text Classification (HP)
100% 6510
8.69 KIPS
 
Text Classification (Q)
93% 6778
9.10 KIPS
 
Machine Translation (SP)
100% 3684
63.5 IPS
 
Machine Translation (HP)
100% 15102
260.1 IPS
 
Machine Translation (Q)
98% 16840
291.0 IPS
 

Compare

Set Baseline Geekbench 6 now available