Geekbench ML Score

6923
Core ML Neural Engine Inference Score
Geekbench ML 0.6.0 for iOS AArch64

Result Information

Upload Date May 29 2024 08:16 AM
Views 1

System Information

System Information
Operating System iOS 17.5.1
Model iPad Pro (12.9-inch 5th generation)
Model IDiPad13,10
Motherboard J523AP
CPU Information
Name Apple M1
Topology 1 Processor, 8 Cores
Identifier ARM
Base Frequency 3.19 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% 1699
318.0 IPS
 
Image Classification (F16)
100% 6163
1.15 KIPS
 
Image Classification (I8)
99% 6402
1.20 KIPS
 
Image Segmentation (F32)
100% 823
13.7 IPS
 
Image Segmentation (F16)
100% 3975
66.4 IPS
 
Image Segmentation (I8)
100% 3945
65.9 IPS
 
Pose Estimation (F32)
100% 16180
19.6 IPS
 
Pose Estimation (F16)
100% 62880
76.1 IPS
 
Pose Estimation (I8)
100% 64678
78.3 IPS
 
Object Detection (F32)
100% 1507
112.5 IPS
 
Object Detection (F16)
100% 3704
276.5 IPS
 
Object Detection (I8)
97% 1470
109.7 IPS
 
Face Detection (F32)
100% 5885
70.0 IPS
 
Face Detection (F16)
97% 11645
138.5 IPS
 
Face Detection (I8)
96% 11785
140.1 IPS
 
Depth Estimation (F32)
100% 12200
94.6 IPS
 
Depth Estimation (F16)
100% 39352
305.2 IPS
 
Depth Estimation (I8)
37% 46121
357.7 IPS
 
Style Transfer (F32)
100% 27305
35.9 IPS
 
Style Transfer (F16)
100% 112363
147.8 IPS
 
Style Transfer (I8)
100% 118715
156.2 IPS
 
Image Super-Resolution (F32)
100% 4925
175.9 IPS
 
Image Super-Resolution (F16)
100% 17940
640.7 IPS
 
Image Super-Resolution (I8)
100% 18014
643.3 IPS
 
Text Classification (F32)
100% 1177
1.69 KIPS
 
Text Classification (F16)
100% 2526
3.63 KIPS
 
Text Classification (I8)
96% 2226
3.20 KIPS
 
Machine Translation (F32)
100% 536
9.87 IPS
 
Machine Translation (F16)
100% 1737
32.0 IPS
 
Machine Translation (I8)
99% 543
10.00 IPS
 

Compare

Set Baseline Geekbench 6 now available