Geekbench AI Score

16224
Single Precision Score
28267
Half Precision Score
13532
Quantized Score
Geekbench AI 1.8.0 for Windows AVX2

Result Information

Upload Date November 20 2026 04:10 PM
Views 4

System Information

AI Information
Framework ONNX
Backend DirectML
Device NVIDIA RTX 4000 SFF Ada Generation
System Information
Operating System Microsoft Windows 11 IoT Enterprise LTSC (64-bit)
Model Velasea W880D4U
Motherboard ASRockRack W880D4U
Power PlanBalanced
CPU Information
Name Intel Core Ultra 9 285K
Topology 1 Processor, 24 Cores
Identifier GenuineIntel Family 6 Model 198 Stepping 2
Base Frequency 3.70 GHz
Cluster 18 Cores
Cluster 216 Cores
Memory Information
Size 32.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 6619
1.23 KIPS
 
Image Classification (HP)
99% 12033
2.24 KIPS
 
Image Classification (Q)
100% 6386
1.19 KIPS
 
Image Segmentation (SP)
100% 16813
272.5 IPS
 
Image Segmentation (HP)
100% 25050
406.1 IPS
 
Image Segmentation (Q)
98% 15128
246.0 IPS
 
Pose Estimation (SP)
100% 86836
101.3 IPS
 
Pose Estimation (HP)
100% 266208
310.6 IPS
 
Pose Estimation (Q)
96% 80004
93.7 IPS
 
Object Detection (SP)
100% 9307
738.2 IPS
 
Object Detection (HP)
100% 17254
1.37 KIPS
 
Object Detection (Q)
85% 7978
642.3 IPS
 
Face Detection (SP)
100% 25976
308.7 IPS
 
Face Detection (HP)
100% 44917
533.7 IPS
 
Face Detection (Q)
97% 21833
260.3 IPS
 
Depth Estimation (SP)
100% 42179
325.0 IPS
 
Depth Estimation (HP)
99% 81999
631.8 IPS
 
Depth Estimation (Q)
78% 34920
278.8 IPS
 
Style Transfer (SP)
100% 146665
188.5 IPS
 
Style Transfer (HP)
100% 393981
506.5 IPS
 
Style Transfer (Q)
98% 129496
167.0 IPS
 
Image Super-Resolution (SP)
100% 25360
936.4 IPS
 
Image Super-Resolution (HP)
100% 38695
1.43 KIPS
 
Image Super-Resolution (Q)
99% 19092
706.9 IPS
 
Text Classification (SP)
100% 1131
1.51 KIPS
 
Text Classification (HP)
100% 1280
1.71 KIPS
 
Text Classification (Q)
97% 813
1.09 KIPS
 
Machine Translation (SP)
100% 3049
52.5 IPS
 
Machine Translation (HP)
100% 3273
56.4 IPS
 
Machine Translation (Q)
70% 2180
40.9 IPS
 

Compare

Set Baseline Geekbench AI now available