Geekbench AI Score

17794
Single Precision Score
19361
Half Precision Score
18139
Quantized Score
Geekbench AI 1.1.0 for macOS AArch64

Result Information

User ___
Upload Date June 19 2025 07:16 AM
Views 6

System Information

AI Information
Framework Core ML
Backend GPU
Device Apple M2 Ultra
System Information
Operating System macOS 15.5 (Build 24F74)
Model Mac Studio (2023)
Model IDMac14,14
Motherboard Mac14,14
CPU Information
Name Apple M2 Ultra
Topology 1 Processor, 24 Cores
Identifier Apple M2 Ultra
Base Frequency 3.30 GHz
Cluster 116 Cores
Cluster 28 Cores
Memory Information
Size 64.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 6246
1.16 KIPS
 
Image Classification (HP)
100% 6299
1.17 KIPS
 
Image Classification (Q)
96% 5824
1.09 KIPS
 
Image Segmentation (SP)
100% 18785
304.5 IPS
 
Image Segmentation (HP)
100% 19574
317.3 IPS
 
Image Segmentation (Q)
99% 18590
301.4 IPS
 
Pose Estimation (SP)
100% 107069
124.9 IPS
 
Pose Estimation (HP)
100% 114588
133.7 IPS
 
Pose Estimation (Q)
96% 109841
128.7 IPS
 
Object Detection (SP)
100% 6267
497.1 IPS
 
Object Detection (HP)
100% 6319
501.2 IPS
 
Object Detection (Q)
92% 5938
474.0 IPS
 
Face Detection (SP)
100% 32214
382.8 IPS
 
Face Detection (HP)
100% 32664
388.1 IPS
 
Face Detection (Q)
100% 28763
341.8 IPS
 
Depth Estimation (SP)
100% 33322
256.7 IPS
 
Depth Estimation (HP)
99% 35550
273.9 IPS
 
Depth Estimation (Q)
86% 32682
255.3 IPS
 
Style Transfer (SP)
100% 192607
247.6 IPS
 
Style Transfer (HP)
100% 201890
259.5 IPS
 
Style Transfer (Q)
96% 196522
253.6 IPS
 
Image Super-Resolution (SP)
100% 24376
900.1 IPS
 
Image Super-Resolution (HP)
100% 25280
933.5 IPS
 
Image Super-Resolution (Q)
99% 22133
817.3 IPS
 
Text Classification (SP)
100% 2338
3.12 KIPS
 
Text Classification (HP)
100% 4031
5.38 KIPS
 
Text Classification (Q)
93% 4023
5.40 KIPS
 
Machine Translation (SP)
100% 3431
59.1 IPS
 
Machine Translation (HP)
100% 3470
59.8 IPS
 
Machine Translation (Q)
98% 3320
57.4 IPS
 

Compare

Set Baseline Geekbench 6 now available