Geekbench AI Score

2782
Single Precision Score
4309
Half Precision Score
3594
Quantized Score
Geekbench AI 1.8.0 for Android AArch64

Result Information

User kristofersodersten
Upload Date October 20 2026 09:50 AM
Views 9

System Information

AI Information
Framework TensorFlow Lite
Backend GPU
Device Qualcomm ARMv8
System Information
Operating System Android 17
Model samsung SM-S948B
Model IDsamsung SM-S948B
Motherboard canoe
Governorwalt
CPU Information
Name Qualcomm ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 81 architecture 8 variant 3 part 2 revision 1
Base Frequency 3.63 GHz
Cluster 16 Cores @ 3.63 GHz
Cluster 22 Cores @ 4.74 GHz
Memory Information
Size 10.82 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 1943
361.3 IPS
 
Image Classification (HP)
100% 3000
557.9 IPS
 
Image Classification (Q)
100% 2296
427.1 IPS
 
Image Segmentation (SP)
100% 2863
46.4 IPS
 
Image Segmentation (HP)
100% 5203
84.4 IPS
 
Image Segmentation (Q)
98% 4479
72.8 IPS
 
Pose Estimation (SP)
100% 12011
14.0 IPS
 
Pose Estimation (HP)
99% 18718
21.8 IPS
 
Pose Estimation (Q)
95% 19175
22.5 IPS
 
Object Detection (SP)
100% 1383
109.7 IPS
 
Object Detection (HP)
99% 1924
152.6 IPS
 
Object Detection (Q)
85% 1840
148.1 IPS
 
Face Detection (SP)
100% 5593
66.5 IPS
 
Face Detection (HP)
100% 10233
121.6 IPS
 
Face Detection (Q)
97% 9182
109.5 IPS
 
Depth Estimation (SP)
100% 6756
52.1 IPS
 
Depth Estimation (HP)
98% 8801
68.0 IPS
 
Depth Estimation (Q)
61% 7441
72.5 IPS
 
Style Transfer (SP)
100% 15723
20.2 IPS
 
Style Transfer (HP)
100% 37672
48.4 IPS
 
Style Transfer (Q)
98% 35414
45.7 IPS
 
Image Super-Resolution (SP)
100% 3895
143.8 IPS
 
Image Super-Resolution (HP)
100% 5398
199.3 IPS
 
Image Super-Resolution (Q)
97% 6362
235.8 IPS
 
Text Classification (SP)
35% 99
967.5 IPS
 
Text Classification (HP)
35% 102
925.9 IPS
 
Text Classification (Q)
33% 79
885.1 IPS
 
Machine Translation (SP)
100% 1316
22.7 IPS
 
Machine Translation (HP)
100% 2104
36.2 IPS
 
Machine Translation (Q)
48% 817
31.3 IPS
 

Compare

Set Baseline Geekbench AI now available