Geekbench AI Score

3569
Single Precision Score
14841
Half Precision Score
14620
Quantized Score
Geekbench AI 1.5.0 for macOS AArch64

Result Information

Upload Date November 11 2025 12:00 AM
Views 1

System Information

AI Information
Framework Core ML
Backend Neural Engine
Device Apple M1 Max
System Information
Operating System macOS 26.1 (Build 25B78)
Model MacBook Pro (16-inch, 2021)
Model IDMacBookPro18,2
Motherboard MacBookPro18,2
CPU Information
Name Apple M1 Max
Topology 1 Processor, 10 Cores
Identifier Apple M1 Max
Base Frequency 3.20 GHz
Cluster 18 Cores
Cluster 22 Cores
Memory Information
Size 32.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 1304
242.5 IPS
 
Image Classification (HP)
100% 6328
1.18 KIPS
 
Image Classification (Q)
96% 6068
1.13 KIPS
 
Image Segmentation (SP)
100% 1513
24.5 IPS
 
Image Segmentation (HP)
100% 8494
137.7 IPS
 
Image Segmentation (Q)
99% 9000
145.9 IPS
 
Pose Estimation (SP)
100% 10303
12.0 IPS
 
Pose Estimation (HP)
100% 70026
81.7 IPS
 
Pose Estimation (Q)
95% 66648
78.1 IPS
 
Object Detection (SP)
100% 1476
117.1 IPS
 
Object Detection (HP)
99% 7129
567.1 IPS
 
Object Detection (Q)
91% 6689
534.4 IPS
 
Face Detection (SP)
100% 3259
38.7 IPS
 
Face Detection (HP)
100% 13248
157.4 IPS
 
Face Detection (Q)
100% 12552
149.1 IPS
 
Depth Estimation (SP)
100% 5576
43.0 IPS
 
Depth Estimation (HP)
99% 47989
369.7 IPS
 
Depth Estimation (Q)
85% 48990
382.8 IPS
 
Style Transfer (SP)
100% 21861
28.1 IPS
 
Style Transfer (HP)
100% 119784
154.0 IPS
 
Style Transfer (Q)
96% 112995
145.8 IPS
 
Image Super-Resolution (SP)
100% 4138
152.8 IPS
 
Image Super-Resolution (HP)
100% 25318
934.9 IPS
 
Image Super-Resolution (Q)
99% 26249
969.3 IPS
 
Text Classification (SP)
100% 2300
3.07 KIPS
 
Text Classification (HP)
99% 4018
5.36 KIPS
 
Text Classification (Q)
93% 3988
5.35 KIPS
 
Machine Translation (SP)
100% 2962
51.0 IPS
 
Machine Translation (HP)
100% 2494
43.0 IPS
 
Machine Translation (Q)
98% 2520
43.5 IPS
 

Compare

Set Baseline Geekbench 6 now available