Geekbench AI Score

573
Single Precision Score
574
Half Precision Score
1301
Quantized Score
Geekbench AI 1.5.0 for Android AArch64

Result Information

Upload Date December 16 2025 11:56 PM
Views 6

System Information

AI Information
Framework TensorFlow Lite
Backend NNAPI
Device Qualcomm ARMv8
System Information
Operating System Android 15
Model Xiaomi 24122RKC7G
Model IDXiaomi 24122RKC7G
Motherboard sun
Governorwalt
CPU Information
Name Qualcomm ARMv8
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 14.76 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 325
60.4 IPS
 
Image Classification (HP)
100% 329
61.1 IPS
 
Image Classification (Q)
99% 865
161.3 IPS
 
Image Segmentation (SP)
100% 427
6.91 IPS
 
Image Segmentation (HP)
100% 427
6.92 IPS
 
Image Segmentation (Q)
98% 828
13.5 IPS
 
Pose Estimation (SP)
100% 603
0.70 IPS
 
Pose Estimation (HP)
100% 603
0.70 IPS
 
Pose Estimation (Q)
98% 2342
2.74 IPS
 
Object Detection (SP)
100% 330
26.2 IPS
 
Object Detection (HP)
100% 332
26.4 IPS
 
Object Detection (Q)
87% 979
78.5 IPS
 
Face Detection (SP)
100% 853
10.1 IPS
 
Face Detection (HP)
100% 855
10.2 IPS
 
Face Detection (Q)
97% 1955
23.3 IPS
 
Depth Estimation (SP)
100% 744
5.73 IPS
 
Depth Estimation (HP)
99% 740
5.70 IPS
 
Depth Estimation (Q)
64% 1770
16.1 IPS
 
Style Transfer (SP)
100% 1333
1.71 IPS
 
Style Transfer (HP)
100% 1328
1.71 IPS
 
Style Transfer (Q)
98% 3584
4.62 IPS
 
Image Super-Resolution (SP)
100% 396
14.6 IPS
 
Image Super-Resolution (HP)
100% 395
14.6 IPS
 
Image Super-Resolution (Q)
97% 1394
51.7 IPS
 
Text Classification (SP)
100% 503
671.8 IPS
 
Text Classification (HP)
100% 504
672.5 IPS
 
Text Classification (Q)
91% 777
1.04 KIPS
 
Machine Translation (SP)
100% 826
14.2 IPS
 
Machine Translation (HP)
100% 837
14.4 IPS
 
Machine Translation (Q)
57% 631
15.4 IPS
 

Compare

Set Baseline Geekbench 6 now available