Geekbench AI Score

21744
Single Precision Score
24252
Half Precision Score
21637
Quantized Score
Geekbench AI 1.2.0 for macOS AArch64

Result Information

Upload Date January 26 2025 03:06 PM
Views 2

System Information

AI Information
Framework Core ML
Backend GPU
Device Apple M4 Max
System Information
Operating System macOS 15.2 (Build 24C101)
Model MacBook Pro (14-inch, 2024)
Model IDMac16,6
Motherboard Mac16,6
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 64.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 10611
1.97 KIPS
 
Image Classification (HP)
100% 10924
2.03 KIPS
 
Image Classification (Q)
97% 9506
1.77 KIPS
 
Image Segmentation (SP)
100% 19243
311.9 IPS
 
Image Segmentation (HP)
100% 20165
326.9 IPS
 
Image Segmentation (Q)
99% 21289
345.1 IPS
 
Pose Estimation (SP)
100% 106642
124.4 IPS
 
Pose Estimation (HP)
100% 109319
127.6 IPS
 
Pose Estimation (Q)
95% 103302
121.1 IPS
 
Object Detection (SP)
100% 8953
710.2 IPS
 
Object Detection (HP)
100% 10445
828.5 IPS
 
Object Detection (Q)
91% 8669
692.7 IPS
 
Face Detection (SP)
100% 39374
467.8 IPS
 
Face Detection (HP)
100% 42568
505.8 IPS
 
Face Detection (Q)
100% 31333
372.3 IPS
 
Depth Estimation (SP)
100% 45433
350.0 IPS
 
Depth Estimation (HP)
99% 47117
363.0 IPS
 
Depth Estimation (Q)
85% 41346
323.1 IPS
 
Style Transfer (SP)
100% 218646
281.1 IPS
 
Style Transfer (HP)
100% 240483
309.1 IPS
 
Style Transfer (Q)
96% 218096
281.4 IPS
 
Image Super-Resolution (SP)
100% 29460
1.09 KIPS
 
Image Super-Resolution (HP)
100% 31049
1.15 KIPS
 
Image Super-Resolution (Q)
99% 25332
935.4 IPS
 
Text Classification (SP)
100% 2961
3.95 KIPS
 
Text Classification (HP)
100% 5350
7.14 KIPS
 
Text Classification (Q)
93% 5330
7.15 KIPS
 
Machine Translation (SP)
100% 3553
61.2 IPS
 
Machine Translation (HP)
100% 3494
60.2 IPS
 
Machine Translation (Q)
98% 3253
56.2 IPS
 

Compare

Set Baseline Geekbench 6 now available