Geekbench AI Score

5398
Single Precision Score
30073
Half Precision Score
33126
Quantized Score
Geekbench AI 1.3.0 for macOS AArch64

Result Information

User idanbo
Upload Date April 04 2025 08:15 AM
Views 9
Notes Mac Studio (2025)/M3 Ultra/32Core CPU/80CoreGPU/32コアNeural Engine/256GB/4TB SSD

System Information

AI Information
Framework Core ML
Backend Neural Engine
Device Apple M3 Ultra
System Information
Operating System macOS 15.4 (Build 24E248)
Model Mac15,14
Model IDMac15,14
Motherboard Mac15,14
CPU Information
Name Apple M3 Ultra
Topology 1 Processor, 32 Cores
Identifier Apple M3 Ultra
Base Frequency 4.04 GHz
Cluster 124 Cores
Cluster 28 Cores
Memory Information
Size 256.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 1720
319.9 IPS
 
Image Classification (HP)
100% 15858
2.95 KIPS
 
Image Classification (Q)
97% 16877
3.15 KIPS
 
Image Segmentation (SP)
100% 2120
34.4 IPS
 
Image Segmentation (HP)
100% 21679
351.4 IPS
 
Image Segmentation (Q)
99% 27396
444.1 IPS
 
Pose Estimation (SP)
100% 22185
25.9 IPS
 
Pose Estimation (HP)
100% 119725
139.7 IPS
 
Pose Estimation (Q)
95% 118301
138.7 IPS
 
Object Detection (SP)
100% 2177
172.7 IPS
 
Object Detection (HP)
100% 17078
1.35 KIPS
 
Object Detection (Q)
91% 17672
1.41 KIPS
 
Face Detection (SP)
100% 4547
54.0 IPS
 
Face Detection (HP)
100% 38534
457.9 IPS
 
Face Detection (Q)
100% 49782
591.5 IPS
 
Depth Estimation (SP)
100% 7937
61.1 IPS
 
Depth Estimation (HP)
99% 106709
822.1 IPS
 
Depth Estimation (Q)
85% 114453
894.8 IPS
 
Style Transfer (SP)
100% 35129
45.2 IPS
 
Style Transfer (HP)
100% 218982
281.5 IPS
 
Style Transfer (Q)
96% 211559
273.0 IPS
 
Image Super-Resolution (SP)
100% 6915
255.3 IPS
 
Image Super-Resolution (HP)
100% 37221
1.37 KIPS
 
Image Super-Resolution (Q)
99% 50069
1.85 KIPS
 
Text Classification (SP)
100% 2794
3.73 KIPS
 
Text Classification (HP)
100% 5070
6.77 KIPS
 
Text Classification (Q)
93% 5032
6.75 KIPS
 
Machine Translation (SP)
100% 4875
84.0 IPS
 
Machine Translation (HP)
100% 5066
87.3 IPS
 
Machine Translation (Q)
98% 5421
93.7 IPS
 

Compare

Set Baseline Geekbench 6 now available