Geekbench ML Score

7364
Core ML Neural Engine Inference Score
Geekbench ML 0.6.0 for iOS AArch64

Result Information

Upload Date May 09 2024 11:35 AM
Views 1

System Information

System Information
Operating System iOS 17.4.1
Model iPad Pro (11-inch, 4th generation)
Model IDiPad14,3
Motherboard J617AP
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 7.33 GB
Inference Information
Framework Core ML
Backend Neural Engine
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 3340
624.9 IPS
 
Image Classification (F16)
100% 6531
1.22 KIPS
 
Image Classification (I8)
99% 7091
1.33 KIPS
 
Image Segmentation (F32)
100% 1436
24.0 IPS
 
Image Segmentation (F16)
100% 1591
26.6 IPS
 
Image Segmentation (I8)
100% 1590
26.6 IPS
 
Pose Estimation (F32)
100% 22593
27.4 IPS
 
Pose Estimation (F16)
100% 81159
98.3 IPS
 
Pose Estimation (I8)
100% 83269
100.8 IPS
 
Object Detection (F32)
100% 2807
209.6 IPS
 
Object Detection (F16)
100% 3629
271.0 IPS
 
Object Detection (I8)
97% 2529
188.8 IPS
 
Face Detection (F32)
100% 9266
110.2 IPS
 
Face Detection (F16)
97% 10620
126.3 IPS
 
Face Detection (I8)
96% 10698
127.2 IPS
 
Depth Estimation (F32)
100% 16021
124.2 IPS
 
Depth Estimation (F16)
100% 36932
286.4 IPS
 
Depth Estimation (I8)
37% 36345
281.9 IPS
 
Style Transfer (F32)
100% 39328
51.7 IPS
 
Style Transfer (F16)
100% 131630
173.1 IPS
 
Style Transfer (I8)
100% 133297
175.3 IPS
 
Image Super-Resolution (F32)
100% 9434
336.9 IPS
 
Image Super-Resolution (F16)
100% 18073
645.4 IPS
 
Image Super-Resolution (I8)
100% 18199
649.9 IPS
 
Text Classification (F32)
100% 1103
1.59 KIPS
 
Text Classification (F16)
100% 743
1.07 KIPS
 
Text Classification (I8)
96% 742
1.07 KIPS
 
Machine Translation (F32)
100% 796
14.7 IPS
 
Machine Translation (F16)
100% 2609
48.0 IPS
 
Machine Translation (I8)
99% 763
14.0 IPS
 

Compare

Set Baseline Geekbench 6 now available