Geekbench AI Score

20456
Single Precision Score
23195
Half Precision Score
21517
Quantized Score
Geekbench AI 1.5.0 for macOS AArch64

Result Information

Upload Date October 26 2025 08:16 PM
Views 2

System Information

AI Information
Framework Core ML
Backend GPU
Device Apple M3 Max
System Information
Operating System macOS 26.0.1 (Build 25A362)
Model MacBook Pro (16-inch, Nov 2023)
Model IDMac15,9
Motherboard Mac15,9
CPU Information
Name Apple M3 Max
Topology 1 Processor, 16 Cores
Identifier Apple M3 Max
Base Frequency 3.99 GHz
Cluster 112 Cores
Cluster 24 Cores
Memory Information
Size 128.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 8982
1.67 KIPS
 
Image Classification (HP)
100% 9859
1.83 KIPS
 
Image Classification (Q)
96% 8203
1.53 KIPS
 
Image Segmentation (SP)
100% 20224
327.8 IPS
 
Image Segmentation (HP)
100% 21037
341.0 IPS
 
Image Segmentation (Q)
99% 20296
329.0 IPS
 
Pose Estimation (SP)
100% 88923
103.8 IPS
 
Pose Estimation (HP)
100% 97530
113.8 IPS
 
Pose Estimation (Q)
96% 93325
109.3 IPS
 
Object Detection (SP)
100% 7832
621.2 IPS
 
Object Detection (HP)
100% 8560
679.0 IPS
 
Object Detection (Q)
92% 8182
653.1 IPS
 
Face Detection (SP)
100% 32622
387.6 IPS
 
Face Detection (HP)
100% 38335
455.5 IPS
 
Face Detection (Q)
100% 31845
378.4 IPS
 
Depth Estimation (SP)
100% 38449
296.2 IPS
 
Depth Estimation (HP)
99% 41985
323.5 IPS
 
Depth Estimation (Q)
86% 37675
294.3 IPS
 
Style Transfer (SP)
100% 186979
240.4 IPS
 
Style Transfer (HP)
100% 212775
273.5 IPS
 
Style Transfer (Q)
96% 204645
264.1 IPS
 
Image Super-Resolution (SP)
100% 27032
998.2 IPS
 
Image Super-Resolution (HP)
100% 26535
979.8 IPS
 
Image Super-Resolution (Q)
99% 24312
897.7 IPS
 
Text Classification (SP)
100% 2978
3.98 KIPS
 
Text Classification (HP)
99% 5179
6.91 KIPS
 
Text Classification (Q)
93% 4995
6.70 KIPS
 
Machine Translation (SP)
100% 5374
92.6 IPS
 
Machine Translation (HP)
100% 5532
95.3 IPS
 
Machine Translation (Q)
98% 5613
97.0 IPS
 

Compare

Set Baseline Geekbench 6 now available