Geekbench AI Score

308
Single Precision Score
500
Half Precision Score
426
Quantized Score
Geekbench AI 1.5.0 for Android AArch64

Result Information

User Oleksandr
Upload Date December 10 2025 04:23 PM
Views 20

System Information

AI Information
Framework TensorFlow Lite
Backend GPU
Device ARM ARMv8
System Information
Operating System Android 15
Model Teclast T65Plus_EEA
Model IDTeclast T65Plus_EEA
Motherboard tb8781_64_cm89_v10
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 4 part 3339 revision 0
Base Frequency 2.00 GHz
Cluster 16 Cores @ 2.00 GHz
Cluster 22 Cores @ 2.20 GHz
Memory Information
Size 7.53 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 182
33.9 IPS
 
Image Classification (HP)
100% 358
66.6 IPS
 
Image Classification (Q)
97% 309
57.7 IPS
 
Image Segmentation (SP)
100% 262
4.24 IPS
 
Image Segmentation (HP)
100% 576
9.33 IPS
 
Image Segmentation (Q)
98% 513
8.34 IPS
 
Pose Estimation (SP)
100% 1037
1.21 IPS
 
Pose Estimation (HP)
100% 1963
2.29 IPS
 
Pose Estimation (Q)
95% 2075
2.43 IPS
 
Object Detection (SP)
100% 160
12.7 IPS
 
Object Detection (HP)
99% 317
25.1 IPS
 
Object Detection (Q)
83% 338
27.3 IPS
 
Face Detection (SP)
100% 598
7.11 IPS
 
Face Detection (HP)
100% 1200
14.3 IPS
 
Face Detection (Q)
96% 1068
12.7 IPS
 
Depth Estimation (SP)
100% 794
6.12 IPS
 
Depth Estimation (HP)
98% 1499
11.6 IPS
 
Depth Estimation (Q)
62% 1204
11.4 IPS
 
Style Transfer (SP)
100% 1952
2.51 IPS
 
Style Transfer (HP)
100% 4052
5.21 IPS
 
Style Transfer (Q)
98% 4129
5.32 IPS
 
Image Super-Resolution (SP)
100% 473
17.5 IPS
 
Image Super-Resolution (HP)
100% 815
30.1 IPS
 
Image Super-Resolution (Q)
97% 745
27.6 IPS
 
Text Classification (SP)
35% 9
84.5 IPS
 
Text Classification (HP)
29% 5
88.3 IPS
 
Text Classification (Q)
32% 7
88.3 IPS
 
Machine Translation (SP)
100% 253
4.35 IPS
 
Machine Translation (HP)
100% 259
4.46 IPS
 
Machine Translation (Q)
43% 64
3.74 IPS
 

Compare

Set Baseline Geekbench 6 now available