Geekbench AI Score

351
Single Precision Score
336
Half Precision Score
582
Quantized Score
Geekbench AI 1.4.0 for Android AArch64

Result Information

User epitgtb29
Upload Date August 24 2025 08:41 PM
Views 11

System Information

AI Information
Framework TensorFlow Lite
Backend CPU
Device ARM MT6785V/CD
System Information
Operating System Android 13
Model Xiaomi M2101K7BNY
Model IDXiaomi M2101K7BNY
Motherboard rosemary
CPU Information
Name ARM MT6785V/CD
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 3 part 3339 revision 0
Base Frequency 2.00 GHz
Cluster 16 Cores @ 2.00 GHz
Cluster 22 Cores @ 2.05 GHz
Memory Information
Size 5.46 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 195
36.2 IPS
 
Image Classification (HP)
100% 194
36.1 IPS
 
Image Classification (Q)
100% 360
66.9 IPS
 
Image Segmentation (SP)
100% 236
3.83 IPS
 
Image Segmentation (HP)
100% 181
2.94 IPS
 
Image Segmentation (Q)
98% 436
7.09 IPS
 
Pose Estimation (SP)
100% 707
0.82 IPS
 
Pose Estimation (HP)
100% 703
0.82 IPS
 
Pose Estimation (Q)
84% 1294
1.54 IPS
 
Object Detection (SP)
100% 192
15.3 IPS
 
Object Detection (HP)
100% 141
11.2 IPS
 
Object Detection (Q)
83% 391
31.6 IPS
 
Face Detection (SP)
100% 472
5.61 IPS
 
Face Detection (HP)
100% 473
5.62 IPS
 
Face Detection (Q)
95% 861
10.3 IPS
 
Depth Estimation (SP)
100% 571
4.40 IPS
 
Depth Estimation (HP)
99% 613
4.73 IPS
 
Depth Estimation (Q)
64% 919
8.42 IPS
 
Style Transfer (SP)
89% 1147
1.49 IPS
 
Style Transfer (HP)
89% 1175
1.53 IPS
 
Style Transfer (Q)
98% 2225
2.87 IPS
 
Image Super-Resolution (SP)
100% 308
11.4 IPS
 
Image Super-Resolution (HP)
100% 318
11.8 IPS
 
Image Super-Resolution (Q)
97% 494
18.3 IPS
 
Text Classification (SP)
100% 168
224.4 IPS
 
Text Classification (HP)
100% 165
220.4 IPS
 
Text Classification (Q)
88% 301
405.5 IPS
 
Machine Translation (SP)
100% 286
4.93 IPS
 
Machine Translation (HP)
100% 299
5.14 IPS
 
Machine Translation (Q)
50% 218
7.75 IPS
 

Compare

Set Baseline Geekbench 6 now available