Geekbench AI Score

6785
Single Precision Score
7070
Half Precision Score
6351
Quantized Score
Geekbench AI 1.0.0 for macOS AArch64

Result Information

User fearlipe
Upload Date August 16 2024 07:10 AM
Views 5

System Information

AI Information
Framework Core ML
Backend GPU
Device Apple M1 Pro
System Information
Operating System macOS 14.6.1 (Build 23G93)
Model MacBook Pro (14-inch, 2021)
Model IDMacBookPro18,3
Motherboard MacBookPro18,3
CPU Information
Name Apple M1 Pro
Topology 1 Processor, 10 Cores
Identifier Apple M1 Pro
Base Frequency 3.04 GHz
Cluster 18 Cores
Cluster 22 Cores
Memory Information
Size 16.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 3464
644.2 IPS
 
Image Classification (HP)
100% 3650
678.8 IPS
 
Image Classification (Q)
99% 3373
629.2 IPS
 
Image Segmentation (SP)
100% 5985
97.0 IPS
 
Image Segmentation (HP)
100% 6102
98.9 IPS
 
Image Segmentation (Q)
100% 5861
95.0 IPS
 
Pose Estimation (SP)
100% 32732
38.2 IPS
 
Pose Estimation (HP)
100% 34152
39.8 IPS
 
Pose Estimation (Q)
96% 30452
35.7 IPS
 
Object Detection (SP)
100% 3594
285.1 IPS
 
Object Detection (HP)
100% 3763
298.5 IPS
 
Object Detection (Q)
92% 2973
237.3 IPS
 
Face Detection (SP)
100% 12170
144.6 IPS
 
Face Detection (HP)
100% 12574
149.4 IPS
 
Face Detection (Q)
100% 8988
106.8 IPS
 
Depth Estimation (SP)
100% 19336
149.0 IPS
 
Depth Estimation (HP)
99% 19708
151.8 IPS
 
Depth Estimation (Q)
97% 20145
155.8 IPS
 
Style Transfer (SP)
100% 52523
67.5 IPS
 
Style Transfer (HP)
100% 62341
80.1 IPS
 
Style Transfer (Q)
100% 62993
81.0 IPS
 
Image Super-Resolution (SP)
100% 10593
391.1 IPS
 
Image Super-Resolution (HP)
100% 11315
417.8 IPS
 
Image Super-Resolution (Q)
98% 8348
309.3 IPS
 
Text Classification (SP)
100% 574
766.2 IPS
 
Text Classification (HP)
100% 545
727.4 IPS
 
Text Classification (Q)
93% 535
717.5 IPS
 
Machine Translation (SP)
100% 1129
19.4 IPS
 
Machine Translation (HP)
100% 1145
19.7 IPS
 
Machine Translation (Q)
98% 1172
20.2 IPS
 

Compare

Set Baseline Geekbench 6 now available