Geekbench AI Score

494
Single Precision Score
537
Half Precision Score
1223
Quantized Score
Geekbench AI 1.5.0 for Android AArch64

Result Information

Upload Date November 26 2025 05:57 AM
Views 4

System Information

AI Information
Framework TensorFlow Lite
Backend NNAPI
Device ARM ARMv8
System Information
Operating System Android 16
Model vivo V2515
Model IDvivo V2515
Motherboard k6993v1_64
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3468 revision 0
Base Frequency 2.70 GHz
Cluster 14 Cores @ 2.70 GHz
Cluster 23 Cores @ 3.50 GHz
Cluster 31 Core @ 4.21 GHz
Memory Information
Size 15.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 359
66.8 IPS
 
Image Classification (HP)
100% 366
68.0 IPS
 
Image Classification (Q)
99% 1025
191.2 IPS
 
Image Segmentation (SP)
100% 479
7.77 IPS
 
Image Segmentation (HP)
100% 482
7.81 IPS
 
Image Segmentation (Q)
98% 980
15.9 IPS
 
Pose Estimation (SP)
100% 632
0.74 IPS
 
Pose Estimation (HP)
100% 603
0.70 IPS
 
Pose Estimation (Q)
98% 2439
2.86 IPS
 
Object Detection (SP)
100% 348
27.6 IPS
 
Object Detection (HP)
100% 350
27.8 IPS
 
Object Detection (Q)
87% 1014
81.3 IPS
 
Face Detection (SP)
100% 800
9.51 IPS
 
Face Detection (HP)
100% 812
9.64 IPS
 
Face Detection (Q)
97% 1993
23.8 IPS
 
Depth Estimation (SP)
100% 721
5.56 IPS
 
Depth Estimation (HP)
99% 623
4.80 IPS
 
Depth Estimation (Q)
64% 1528
13.9 IPS
 
Style Transfer (SP)
100% 839
1.08 IPS
 
Style Transfer (HP)
100% 918
1.18 IPS
 
Style Transfer (Q)
98% 2359
3.04 IPS
 
Image Super-Resolution (SP)
100% 367
13.6 IPS
 
Image Super-Resolution (HP)
100% 365
13.5 IPS
 
Image Super-Resolution (Q)
97% 1352
50.1 IPS
 
Text Classification (SP)
100% 206
274.8 IPS
 
Text Classification (HP)
100% 496
662.1 IPS
 
Text Classification (Q)
91% 674
905.7 IPS
 
Machine Translation (SP)
100% 638
11.0 IPS
 
Machine Translation (HP)
100% 649
11.2 IPS
 
Machine Translation (Q)
57% 462
11.3 IPS
 

Compare

Set Baseline Geekbench 6 now available