Geekbench ML Score

7430
Core ML Neural Engine Inference Score
Geekbench ML 0.6.0 for iOS AArch64

Result Information

Upload Date July 18 2024 02:41 PM
Views 1

System Information

System Information
Operating System iOS 17.6
Model iPad16,5
Model IDiPad16,5
Motherboard J720AP
CPU Information
Name ARM
Topology 1 Processor, 9 Cores
Identifier ARM
Base Frequency 4.40 GHz
Cluster 13 Cores
Cluster 26 Cores
L1 Instruction Cache128 KB x 1
L1 Data Cache64.0 KB x 1
L2 Cache4.00 MB x 1
Memory Information
Size 7.20 GB
Inference Information
Framework Core ML
Backend Neural Engine
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 4056
759.0 IPS
 
Image Classification (F16)
100% 8765
1.64 KIPS
 
Image Classification (I8)
99% 9796
1.83 KIPS
 
Image Segmentation (F32)
100% 1791
29.9 IPS
 
Image Segmentation (F16)
100% 2266
37.9 IPS
 
Image Segmentation (I8)
100% 1833
30.6 IPS
 
Pose Estimation (F32)
100% 25937
31.4 IPS
 
Pose Estimation (F16)
100% 98506
119.3 IPS
 
Pose Estimation (I8)
100% 101892
123.4 IPS
 
Object Detection (F32)
100% 3410
254.6 IPS
 
Object Detection (F16)
100% 5884
439.3 IPS
 
Object Detection (I8)
97% 3360
250.8 IPS
 
Face Detection (F32)
100% 11521
137.0 IPS
 
Face Detection (F16)
97% 14527
172.7 IPS
 
Face Detection (I8)
96% 14673
174.5 IPS
 
Depth Estimation (F32)
100% 19934
154.6 IPS
 
Depth Estimation (F16)
100% 49101
380.8 IPS
 
Depth Estimation (I8)
37% 51645
400.5 IPS
 
Style Transfer (F32)
100% 56490
74.3 IPS
 
Style Transfer (F16)
100% 152472
200.6 IPS
 
Style Transfer (I8)
100% 153377
201.8 IPS
 
Image Super-Resolution (F32)
100% 8703
310.8 IPS
 
Image Super-Resolution (F16)
100% 21844
780.1 IPS
 
Image Super-Resolution (I8)
100% 22182
792.2 IPS
 
Text Classification (F32)
100% 1392
2.00 KIPS
 
Text Classification (F16)
100% 1459
2.10 KIPS
 
Text Classification (I8)
96% 1056
1.52 KIPS
 
Machine Translation (F32)
100% 1123
20.7 IPS
 
Machine Translation (F16)
100% 1787
32.9 IPS
 
Machine Translation (I8)
0
 

Compare

Set Baseline Geekbench 6 now available