Geekbench ML Score

1143
Core ML Neural Engine Inference Score
Geekbench ML 0.6.0 for macOS AArch64

Result Information

Upload Date September 11 2024 12:19 PM
Views 2

System Information

System Information
Operating System macOS 14.4.1 (Build 23E224)
Model VirtualMac2,1
Model IDVirtualMac2,1
Motherboard VirtualMac2,1
CPU Information
Name Apple M2 Pro (Virtual)
Topology 1 Processor, 8 Cores
Identifier Apple M2 Pro (Virtual)
Base Frequency 3.47 GHz
Cluster 18 Cores
L1 Instruction Cache192 KB x 1
L1 Data Cache128 KB x 1
L2 Cache16.0 MB x 1
Memory Information
Size 28.00 GB
Inference Information
Framework Core ML
Backend Neural Engine
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 566
105.9 IPS
 
Image Classification (F16)
100% 516
96.5 IPS
 
Image Classification (I8)
99% 509
95.3 IPS
 
Image Segmentation (F32)
100% 514
8.59 IPS
 
Image Segmentation (F16)
100% 490
8.18 IPS
 
Image Segmentation (I8)
100% 493
8.23 IPS
 
Pose Estimation (F32)
100% 2958
3.58 IPS
 
Pose Estimation (F16)
100% 4529
5.48 IPS
 
Pose Estimation (I8)
100% 3022
3.66 IPS
 
Object Detection (F32)
100% 679
50.7 IPS
 
Object Detection (F16)
100% 603
45.0 IPS
 
Object Detection (I8)
97% 657
49.1 IPS
 
Face Detection (F32)
100% 1588
18.9 IPS
 
Face Detection (F16)
100% 1512
18.0 IPS
 
Face Detection (I8)
99% 1562
18.6 IPS
 
Depth Estimation (F32)
100% 2125
16.5 IPS
 
Depth Estimation (F16)
100% 1971
15.3 IPS
 
Depth Estimation (I8)
99% 1958
15.2 IPS
 
Style Transfer (F32)
100% 7584
9.98 IPS
 
Style Transfer (F16)
100% 7347
9.66 IPS
 
Style Transfer (I8)
100% 7363
9.69 IPS
 
Image Super-Resolution (F32)
100% 1741
62.2 IPS
 
Image Super-Resolution (F16)
100% 1734
61.9 IPS
 
Image Super-Resolution (I8)
100% 1737
62.0 IPS
 
Text Classification (F32)
100% 201
288.5 IPS
 
Text Classification (F16)
100% 360
517.4 IPS
 
Text Classification (I8)
96% 364
523.3 IPS
 
Machine Translation (F32)
100% 591
10.9 IPS
 
Machine Translation (F16)
100% 465
8.56 IPS
 
Machine Translation (I8)
99% 583
10.7 IPS
 

Compare

Set Baseline Geekbench 6 now available