Geekbench ML Score

7455
Core ML Neural Engine Inference Score
Geekbench ML 0.6.0 for iOS AArch64

Result Information

Upload Date May 09 2024 05:36 AM
Views 2

System Information

System Information
Operating System iOS 17.5
Model iPad Pro (12.9-inch, 6th generation)
Model IDiPad14,5
Motherboard J620AP
CPU Information
Name Apple M2
Topology 1 Processor, 8 Cores
Identifier ARM
Base Frequency 3.49 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 7.31 GB
Inference Information
Framework Core ML
Backend Neural Engine
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 3320
621.4 IPS
 
Image Classification (F16)
100% 6529
1.22 KIPS
 
Image Classification (I8)
99% 7035
1.32 KIPS
 
Image Segmentation (F32)
100% 1447
24.2 IPS
 
Image Segmentation (F16)
100% 1591
26.6 IPS
 
Image Segmentation (I8)
100% 1583
26.4 IPS
 
Pose Estimation (F32)
100% 22705
27.5 IPS
 
Pose Estimation (F16)
100% 83037
100.6 IPS
 
Pose Estimation (I8)
100% 83936
101.6 IPS
 
Object Detection (F32)
100% 2832
211.4 IPS
 
Object Detection (F16)
100% 3989
297.8 IPS
 
Object Detection (I8)
97% 2612
195.0 IPS
 
Face Detection (F32)
100% 9287
110.4 IPS
 
Face Detection (F16)
97% 10634
126.4 IPS
 
Face Detection (I8)
96% 10718
127.4 IPS
 
Depth Estimation (F32)
100% 16067
124.6 IPS
 
Depth Estimation (F16)
100% 36930
286.4 IPS
 
Depth Estimation (I8)
37% 44283
343.4 IPS
 
Style Transfer (F32)
100% 39280
51.7 IPS
 
Style Transfer (F16)
100% 132396
174.2 IPS
 
Style Transfer (I8)
100% 132121
173.8 IPS
 
Image Super-Resolution (F32)
100% 9352
334.0 IPS
 
Image Super-Resolution (F16)
100% 18590
663.9 IPS
 
Image Super-Resolution (I8)
100% 17138
612.1 IPS
 
Text Classification (F32)
100% 1136
1.63 KIPS
 
Text Classification (F16)
100% 762
1.10 KIPS
 
Text Classification (I8)
96% 761
1.09 KIPS
 
Machine Translation (F32)
100% 794
14.6 IPS
 
Machine Translation (F16)
100% 2595
47.8 IPS
 
Machine Translation (I8)
99% 744
13.7 IPS
 

Compare

Set Baseline Geekbench 6 now available