Geekbench AI Score

12655
Single Precision Score
14656
Half Precision Score
12571
Quantized Score
Geekbench AI 1.2.0 for macOS AArch64

Result Information

User Simondeminuit
Upload Date January 16 2025 11:59 PM
Views 13

System Information

AI Information
Framework Core ML
Backend GPU
Device Apple M4 Pro
System Information
Operating System macOS 15.2 (Build 24C101)
Model Mac mini (2024)
Model IDMac16,11
Motherboard Mac16,11
CPU Information
Name Apple M4 Pro
Topology 1 Processor, 12 Cores
Identifier Apple M4 Pro
Base Frequency 4.50 GHz
Cluster 18 Cores
Cluster 24 Cores
Memory Information
Size 48.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 6836
1.27 KIPS
 
Image Classification (HP)
100% 7295
1.36 KIPS
 
Image Classification (Q)
97% 6218
1.16 KIPS
 
Image Segmentation (SP)
100% 10574
171.4 IPS
 
Image Segmentation (HP)
100% 11016
178.6 IPS
 
Image Segmentation (Q)
99% 11254
182.4 IPS
 
Pose Estimation (SP)
100% 46231
53.9 IPS
 
Pose Estimation (HP)
100% 48309
56.4 IPS
 
Pose Estimation (Q)
95% 45737
53.6 IPS
 
Object Detection (SP)
100% 6681
530.0 IPS
 
Object Detection (HP)
100% 7226
573.1 IPS
 
Object Detection (Q)
91% 6053
483.7 IPS
 
Face Detection (SP)
100% 22619
268.8 IPS
 
Face Detection (HP)
100% 25426
302.1 IPS
 
Face Detection (Q)
100% 18343
218.0 IPS
 
Depth Estimation (SP)
100% 30871
237.8 IPS
 
Depth Estimation (HP)
99% 32797
252.7 IPS
 
Depth Estimation (Q)
85% 27262
213.1 IPS
 
Style Transfer (SP)
100% 104672
134.6 IPS
 
Style Transfer (HP)
100% 112447
144.6 IPS
 
Style Transfer (Q)
96% 106992
138.1 IPS
 
Image Super-Resolution (SP)
100% 16438
607.0 IPS
 
Image Super-Resolution (HP)
100% 17241
636.6 IPS
 
Image Super-Resolution (Q)
99% 13734
507.1 IPS
 
Text Classification (SP)
100% 2963
3.95 KIPS
 
Text Classification (HP)
100% 5200
6.94 KIPS
 
Text Classification (Q)
93% 5140
6.90 KIPS
 
Machine Translation (SP)
100% 1326
22.8 IPS
 
Machine Translation (HP)
100% 1940
33.4 IPS
 
Machine Translation (Q)
98% 1347
23.3 IPS
 

Compare

Set Baseline Geekbench 6 now available