Geekbench ML Score

6798
Core ML Neural Engine Inference Score
Geekbench ML 0.6.0 for iOS AArch64

Result Information

Upload Date September 02 2024 07:02 PM
Views 1

System Information

System Information
Operating System iOS 18.0
Model iPad Pro (11-inch 3rd generation)
Model IDiPad13,4
Motherboard J517AP
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 7.37 GB
Inference Information
Framework Core ML
Backend Neural Engine
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 1835
343.3 IPS
 
Image Classification (F16)
100% 5684
1.06 KIPS
 
Image Classification (I8)
99% 6338
1.19 KIPS
 
Image Segmentation (F32)
100% 804
13.4 IPS
 
Image Segmentation (F16)
100% 1095
18.3 IPS
 
Image Segmentation (I8)
100% 1092
18.2 IPS
 
Pose Estimation (F32)
100% 16388
19.8 IPS
 
Pose Estimation (F16)
100% 62975
76.3 IPS
 
Pose Estimation (I8)
100% 64793
78.5 IPS
 
Object Detection (F32)
100% 1527
114.0 IPS
 
Object Detection (F16)
100% 4046
302.1 IPS
 
Object Detection (I8)
97% 1913
142.9 IPS
 
Face Detection (F32)
100% 6444
76.6 IPS
 
Face Detection (F16)
97% 11809
140.4 IPS
 
Face Detection (I8)
96% 11821
140.6 IPS
 
Depth Estimation (F32)
100% 12114
93.9 IPS
 
Depth Estimation (F16)
100% 42120
326.6 IPS
 
Depth Estimation (I8)
37% 43909
340.5 IPS
 
Style Transfer (F32)
100% 24304
32.0 IPS
 
Style Transfer (F16)
100% 112600
148.1 IPS
 
Style Transfer (I8)
100% 118817
156.3 IPS
 
Image Super-Resolution (F32)
100% 4377
156.3 IPS
 
Image Super-Resolution (F16)
100% 23475
838.4 IPS
 
Image Super-Resolution (I8)
100% 23865
852.3 IPS
 
Text Classification (F32)
100% 1993
2.86 KIPS
 
Text Classification (F16)
100% 3452
4.96 KIPS
 
Text Classification (I8)
96% 3248
4.67 KIPS
 
Machine Translation (F32)
100% 628
11.6 IPS
 
Machine Translation (F16)
100% 1618
29.8 IPS
 
Machine Translation (I8)
99% 511
9.41 IPS
 

Compare

Set Baseline Geekbench 6 now available