Geekbench AI Score

4721
Single Precision Score
36368
Half Precision Score
50701
Quantized Score
Geekbench AI 1.2.0 for macOS AArch64

Result Information

User oscarbg81
Upload Date December 07 2024 07:00 AM
Views 12

System Information

AI Information
Framework Core ML
Backend Neural Engine
Device Apple M4
System Information
Operating System macOS 15.2 (Build 24C5089c)
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 4.41 GHz
Cluster 14 Cores
Cluster 26 Cores
Memory Information
Size 16.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 2041
379.5 IPS
 
Image Classification (HP)
100% 18642
3.47 KIPS
 
Image Classification (Q)
97% 23560
4.40 KIPS
 
Image Segmentation (SP)
100% 2090
33.9 IPS
 
Image Segmentation (HP)
100% 21850
354.2 IPS
 
Image Segmentation (Q)
99% 30284
490.9 IPS
 
Pose Estimation (SP)
100% 9982
11.6 IPS
 
Pose Estimation (HP)
100% 123731
144.4 IPS
 
Pose Estimation (Q)
95% 232538
272.5 IPS
 
Object Detection (SP)
100% 2301
182.5 IPS
 
Object Detection (HP)
100% 17836
1.41 KIPS
 
Object Detection (Q)
91% 25261
2.02 KIPS
 
Face Detection (SP)
100% 4010
47.6 IPS
 
Face Detection (HP)
100% 42893
509.7 IPS
 
Face Detection (Q)
100% 64825
770.3 IPS
 
Depth Estimation (SP)
100% 8330
64.2 IPS
 
Depth Estimation (HP)
99% 100052
770.8 IPS
 
Depth Estimation (Q)
85% 167074
1.31 KIPS
 
Style Transfer (SP)
100% 25360
32.6 IPS
 
Style Transfer (HP)
100% 234108
300.9 IPS
 
Style Transfer (Q)
96% 298474
385.2 IPS
 
Image Super-Resolution (SP)
100% 5779
213.4 IPS
 
Image Super-Resolution (HP)
100% 49506
1.83 KIPS
 
Image Super-Resolution (Q)
99% 83268
3.07 KIPS
 
Text Classification (SP)
100% 2949
3.94 KIPS
 
Text Classification (HP)
100% 5607
7.48 KIPS
 
Text Classification (Q)
93% 5613
7.53 KIPS
 
Machine Translation (SP)
100% 3893
67.1 IPS
 
Machine Translation (HP)
100% 16148
278.2 IPS
 
Machine Translation (Q)
98% 17727
306.3 IPS
 

Compare

Set Baseline Geekbench 6 now available