Geekbench ML Score

5649
Core ML GPU Inference Score
Geekbench ML 0.6.0 for macOS AArch64

Result Information

Upload Date July 05 2025 12:43 PM
Views 3

System Information

System Information
Operating System macOS 16.0 (Build 25A5295e)
Model MacBook Pro (13-inch, 2022)
Model IDMac14,7
Motherboard Mac14,7
CPU Information
Name Apple M2
Topology 1 Processor, 8 Cores
Identifier Apple M2
Base Frequency 3.47 GHz
Cluster 14 Cores
Cluster 24 Cores
L1 Instruction Cache128 KB x 1
L1 Data Cache64.0 KB x 1
L2 Cache4.00 MB x 1
Memory Information
Size 8.00 GB
Inference Information
Framework Core ML
Backend GPU
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 3414
638.8 IPS
 
Image Classification (F16)
100% 3449
645.5 IPS
 
Image Classification (I8)
99% 3415
639.0 IPS
 
Image Segmentation (F32)
100% 1472
24.6 IPS
 
Image Segmentation (F16)
100% 1216
20.3 IPS
 
Image Segmentation (I8)
100% 1436
24.0 IPS
 
Pose Estimation (F32)
100% 23024
27.9 IPS
 
Pose Estimation (F16)
100% 25636
31.0 IPS
 
Pose Estimation (I8)
100% 23375
28.3 IPS
 
Object Detection (F32)
100% 2783
207.8 IPS
 
Object Detection (F16)
100% 2912
217.4 IPS
 
Object Detection (I8)
97% 2768
206.6 IPS
 
Face Detection (F32)
100% 7194
85.5 IPS
 
Face Detection (F16)
98% 5315
63.2 IPS
 
Face Detection (I8)
97% 10713
127.4 IPS
 
Depth Estimation (F32)
100% 16133
125.1 IPS
 
Depth Estimation (F16)
100% 17656
136.9 IPS
 
Depth Estimation (I8)
99% 18131
140.6 IPS
 
Style Transfer (F32)
100% 40988
53.9 IPS
 
Style Transfer (F16)
100% 47957
63.1 IPS
 
Style Transfer (I8)
100% 32540
42.8 IPS
 
Image Super-Resolution (F32)
100% 6984
249.4 IPS
 
Image Super-Resolution (F16)
100% 10445
373.0 IPS
 
Image Super-Resolution (I8)
100% 7689
274.6 IPS
 
Text Classification (F32)
100% 2075
2.98 KIPS
 
Text Classification (F16)
100% 3290
4.73 KIPS
 
Text Classification (I8)
96% 3342
4.80 KIPS
 
Machine Translation (F32)
100% 870
16.0 IPS
 
Machine Translation (F16)
100% 863
15.9 IPS
 
Machine Translation (I8)
99% 874
16.1 IPS
 

Compare

Set Baseline Geekbench 6 now available