Geekbench AI Score

31685
Single Precision Score
43790
Half Precision Score
28088
Quantized Score
Geekbench AI 1.5.0 for Windows AVX2

Result Information

User nwgat
Upload Date May 29 2026 08:14 PM
Views 13

System Information

AI Information
Framework ONNX
Backend DirectML
Device AMD Radeon RX 9070 XT
System Information
Operating System Microsoft Windows 11 Pro (64-bit)
Model Gigabyte Technology Co., Ltd. B550I AORUS PRO AX
Motherboard Gigabyte Technology Co., Ltd. B550I AORUS PRO AX
Power PlanHigh performance
CPU Information
Name AMD Ryzen 9 5900X
Topology 1 Processor, 12 Cores, 24 Threads
Identifier AuthenticAMD Family 25 Model 33 Stepping 0
Base Frequency 3.70 GHz
Cluster 112 Cores
Memory Information
Size 32.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 13314
2.48 KIPS
 
Image Classification (HP)
100% 15277
2.84 KIPS
 
Image Classification (Q)
99% 13289
2.48 KIPS
 
Image Segmentation (SP)
100% 17763
288.0 IPS
 
Image Segmentation (HP)
100% 21064
341.5 IPS
 
Image Segmentation (Q)
98% 21126
343.5 IPS
 
Pose Estimation (SP)
100% 203478
237.4 IPS
 
Pose Estimation (HP)
100% 607686
709.1 IPS
 
Pose Estimation (Q)
96% 184588
216.2 IPS
 
Object Detection (SP)
100% 15760
1.25 KIPS
 
Object Detection (HP)
99% 18407
1.46 KIPS
 
Object Detection (Q)
87% 14496
1.16 KIPS
 
Face Detection (SP)
100% 43671
518.9 IPS
 
Face Detection (HP)
100% 52818
627.6 IPS
 
Face Detection (Q)
97% 41011
489.0 IPS
 
Depth Estimation (SP)
100% 82957
639.1 IPS
 
Depth Estimation (HP)
99% 123084
948.3 IPS
 
Depth Estimation (Q)
77% 69553
557.6 IPS
 
Style Transfer (SP)
100% 376357
483.8 IPS
 
Style Transfer (HP)
100% 676569
869.7 IPS
 
Style Transfer (Q)
98% 352254
454.2 IPS
 
Image Super-Resolution (SP)
100% 48949
1.81 KIPS
 
Image Super-Resolution (HP)
100% 75814
2.80 KIPS
 
Image Super-Resolution (Q)
99% 41483
1.54 KIPS
 
Text Classification (SP)
100% 3602
4.81 KIPS
 
Text Classification (HP)
100% 3844
5.13 KIPS
 
Text Classification (Q)
97% 2822
3.78 KIPS
 
Machine Translation (SP)
100% 5594
96.4 IPS
 
Machine Translation (HP)
100% 5620
96.8 IPS
 
Machine Translation (Q)
70% 3460
64.9 IPS
 

Compare

Set Baseline Geekbench AI now available