Geekbench AI Score

26294
Single Precision Score
33761
Half Precision Score
20877
Quantized Score
Geekbench AI 1.0.0 for Windows AVX2

Result Information

User ducphuc
Upload Date August 16 2024 03:46 PM
Views 4

System Information

AI Information
Framework ONNX
Backend DirectML
Device NVIDIA GeForce RTX 4090
System Information
Operating System Microsoft Windows 11 Enterprise (64-bit)
Model LENOVO 30E0CTO1WW
Motherboard LENOVO 1046
Power PlanAMD Ryzen Balanced
CPU Information
Name AMD Ryzen Threadripper PRO 3975WX
Topology 1 Processor, 32 Cores, 64 Threads
Identifier AuthenticAMD Family 23 Model 49 Stepping 0
Base Frequency 3.50 GHz
Cluster 132 Cores
Memory Information
Size 128.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 10925
2.03 KIPS
 
Image Classification (HP)
99% 13808
2.58 KIPS
 
Image Classification (Q)
100% 9409
1.75 KIPS
 
Image Segmentation (SP)
100% 10554
171.1 IPS
 
Image Segmentation (HP)
100% 11574
187.6 IPS
 
Image Segmentation (Q)
98% 10108
164.4 IPS
 
Pose Estimation (SP)
100% 277099
323.3 IPS
 
Pose Estimation (HP)
100% 447532
522.2 IPS
 
Pose Estimation (Q)
97% 243676
285.3 IPS
 
Object Detection (SP)
100% 13074
1.04 KIPS
 
Object Detection (HP)
100% 15934
1.26 KIPS
 
Object Detection (Q)
89% 11582
927.1 IPS
 
Face Detection (SP)
100% 34545
410.5 IPS
 
Face Detection (HP)
100% 41212
489.7 IPS
 
Face Detection (Q)
97% 29782
355.1 IPS
 
Depth Estimation (SP)
100% 80779
622.4 IPS
 
Depth Estimation (HP)
99% 127221
980.2 IPS
 
Depth Estimation (Q)
75% 60085
485.0 IPS
 
Style Transfer (SP)
100% 448817
577.0 IPS
 
Style Transfer (HP)
100% 744786
957.4 IPS
 
Style Transfer (Q)
98% 394152
508.3 IPS
 
Image Super-Resolution (SP)
100% 42247
1.56 KIPS
 
Image Super-Resolution (HP)
100% 51666
1.91 KIPS
 
Image Super-Resolution (Q)
99% 33288
1.23 KIPS
 
Text Classification (SP)
100% 2861
3.82 KIPS
 
Text Classification (HP)
99% 3382
4.51 KIPS
 
Text Classification (Q)
97% 1853
2.48 KIPS
 
Machine Translation (SP)
100% 2499
43.1 IPS
 
Machine Translation (HP)
100% 2474
42.6 IPS
 
Machine Translation (Q)
60% 1347
30.0 IPS
 

Compare

Set Baseline Geekbench 6 now available