Geekbench AI Score

5606
Single Precision Score
36556
Half Precision Score
49472
Quantized Score
Geekbench AI 1.4.0 for macOS AArch64

Result Information

User Ronlee97
Upload Date August 06 2025 04:33 PM
Views 10

System Information

AI Information
Framework Core ML
Backend Neural Engine
Device Apple M4 Max
System Information
Operating System macOS 15.6 (Build 24G84)
Model Mac16,9
Model IDMac16,9
Motherboard Mac16,9
CPU Information
Name Apple M4 Max
Topology 1 Processor, 16 Cores
Identifier Apple M4 Max
Base Frequency 4.51 GHz
Cluster 112 Cores
Cluster 24 Cores
Memory Information
Size 128.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 2131
396.3 IPS
 
Image Classification (HP)
100% 17198
3.20 KIPS
 
Image Classification (Q)
97% 22151
4.13 KIPS
 
Image Segmentation (SP)
100% 2351
38.1 IPS
 
Image Segmentation (HP)
100% 22658
367.3 IPS
 
Image Segmentation (Q)
99% 31444
509.7 IPS
 
Pose Estimation (SP)
100% 17975
21.0 IPS
 
Pose Estimation (HP)
100% 122509
142.9 IPS
 
Pose Estimation (Q)
95% 229755
269.3 IPS
 
Object Detection (SP)
100% 2623
208.1 IPS
 
Object Detection (HP)
100% 19293
1.53 KIPS
 
Object Detection (Q)
91% 23356
1.87 KIPS
 
Face Detection (SP)
100% 4960
58.9 IPS
 
Face Detection (HP)
100% 45195
537.0 IPS
 
Face Detection (Q)
100% 64778
769.7 IPS
 
Depth Estimation (SP)
100% 8880
68.4 IPS
 
Depth Estimation (HP)
99% 109602
844.4 IPS
 
Depth Estimation (Q)
85% 176152
1.38 KIPS
 
Style Transfer (SP)
100% 26192
33.7 IPS
 
Style Transfer (HP)
100% 231854
298.1 IPS
 
Style Transfer (Q)
96% 297295
383.6 IPS
 
Image Super-Resolution (SP)
100% 6071
224.2 IPS
 
Image Super-Resolution (HP)
100% 45930
1.70 KIPS
 
Image Super-Resolution (Q)
99% 77046
2.84 KIPS
 
Text Classification (SP)
100% 2983
3.98 KIPS
 
Text Classification (HP)
100% 5381
7.18 KIPS
 
Text Classification (Q)
93% 5486
7.36 KIPS
 
Machine Translation (SP)
100% 6215
107.1 IPS
 
Machine Translation (HP)
100% 16305
280.9 IPS
 
Machine Translation (Q)
98% 16390
283.2 IPS
 

Compare

Set Baseline Geekbench 6 now available