Geekbench AI Score

5095
Single Precision Score
36650
Half Precision Score
49650
Quantized Score
Geekbench AI 1.5.0 for iOS AArch64

Result Information

User dr904
Upload Date October 07 2025 12:21 AM
Views 6
Notes * App is not optimized for iPhone 17 Pro.

System Information

AI Information
Framework Core ML
Backend Neural Engine
Device ARM
System Information
Operating System iOS 26.0.1
Model iPhone18,1
Model IDiPhone18,1
Motherboard V53AP
CPU Information
Name ARM
Topology 1 Processor, 6 Cores
Identifier ARM
Base Frequency 4.26 GHz
Cluster 12 Cores
Cluster 24 Cores
Memory Information
Size 11.45 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 2336
434.4 IPS
 
Image Classification (HP)
100% 16149
3.00 KIPS
 
Image Classification (Q)
97% 22384
4.18 KIPS
 
Image Segmentation (SP)
100% 2098
34.0 IPS
 
Image Segmentation (HP)
100% 20337
329.7 IPS
 
Image Segmentation (Q)
99% 25847
419.0 IPS
 
Pose Estimation (SP)
100% 10274
12.0 IPS
 
Pose Estimation (HP)
100% 157481
183.8 IPS
 
Pose Estimation (Q)
95% 288828
338.5 IPS
 
Object Detection (SP)
100% 2616
207.5 IPS
 
Object Detection (HP)
100% 17919
1.42 KIPS
 
Object Detection (Q)
91% 25226
2.02 KIPS
 
Face Detection (SP)
100% 4555
54.1 IPS
 
Face Detection (HP)
100% 40853
485.4 IPS
 
Face Detection (Q)
100% 55342
657.6 IPS
 
Depth Estimation (SP)
100% 8370
64.5 IPS
 
Depth Estimation (HP)
99% 105198
810.5 IPS
 
Depth Estimation (Q)
85% 163660
1.28 KIPS
 
Style Transfer (SP)
100% 24525
31.5 IPS
 
Style Transfer (HP)
100% 257816
331.4 IPS
 
Style Transfer (Q)
96% 351433
453.5 IPS
 
Image Super-Resolution (SP)
100% 5410
199.8 IPS
 
Image Super-Resolution (HP)
100% 59698
2.20 KIPS
 
Image Super-Resolution (Q)
99% 84792
3.13 KIPS
 
Text Classification (SP)
100% 3544
4.73 KIPS
 
Text Classification (HP)
99% 6413
8.56 KIPS
 
Text Classification (Q)
93% 6387
8.57 KIPS
 
Machine Translation (SP)
100% 4992
86.0 IPS
 
Machine Translation (HP)
100% 11125
191.6 IPS
 
Machine Translation (Q)
98% 12530
216.5 IPS
 

Compare

Set Baseline Geekbench 6 now available