Geekbench AI Score

1057
Single Precision Score
1040
Half Precision Score
1688
Quantized Score
Geekbench AI 1.4.0 for Android AArch64

Result Information

Upload Date August 09 2025 03:55 PM
Views 4

System Information

AI Information
Framework TensorFlow Lite
Backend CPU
Device Qualcomm SM8750
System Information
Operating System Android 15
Model HONOR MBH-N49
Model IDHONOR MBH-N49
Motherboard MBH
Governorwalt
CPU Information
Name Qualcomm SM8750
Topology 1 Processor, 8 Cores
Identifier ARM implementer 81 architecture 8 variant 3 part 1 revision 4
Base Frequency 3.53 GHz
Cluster 16 Cores @ 3.53 GHz
Cluster 22 Cores @ 4.32 GHz
Memory Information
Size 14.89 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 733
136.3 IPS
 
Image Classification (HP)
100% 652
121.3 IPS
 
Image Classification (Q)
100% 1205
224.1 IPS
 
Image Segmentation (SP)
100% 893
14.5 IPS
 
Image Segmentation (HP)
100% 869
14.1 IPS
 
Image Segmentation (Q)
98% 1837
29.9 IPS
 
Pose Estimation (SP)
100% 2050
2.39 IPS
 
Pose Estimation (HP)
100% 2046
2.39 IPS
 
Pose Estimation (Q)
84% 4875
5.78 IPS
 
Object Detection (SP)
98% 626
49.8 IPS
 
Object Detection (HP)
98% 645
51.3 IPS
 
Object Detection (Q)
83% 1365
110.3 IPS
 
Face Detection (SP)
100% 2006
23.8 IPS
 
Face Detection (HP)
100% 1958
23.3 IPS
 
Face Detection (Q)
95% 2849
34.0 IPS
 
Depth Estimation (SP)
99% 1747
13.5 IPS
 
Depth Estimation (HP)
99% 1745
13.5 IPS
 
Depth Estimation (Q)
64% 3369
30.9 IPS
 
Style Transfer (SP)
89% 4529
5.88 IPS
 
Style Transfer (HP)
89% 4605
5.98 IPS
 
Style Transfer (Q)
98% 8780
11.3 IPS
 
Image Super-Resolution (SP)
100% 953
35.2 IPS
 
Image Super-Resolution (HP)
100% 943
34.8 IPS
 
Image Super-Resolution (Q)
97% 2099
77.8 IPS
 
Text Classification (SP)
100% 234
312.2 IPS
 
Text Classification (HP)
99% 242
323.5 IPS
 
Text Classification (Q)
88% 267
359.5 IPS
 
Machine Translation (SP)
100% 591
10.2 IPS
 
Machine Translation (HP)
100% 556
9.58 IPS
 
Machine Translation (Q)
50% 271
9.60 IPS
 

Compare

Set Baseline Geekbench 6 now available