Geekbench AI Score

2866
Single Precision Score
2865
Half Precision Score
2451
Quantized Score
Geekbench AI 1.5.0 for Linux AVX2

Result Information

User jinpachu
Upload Date November 09 2025 01:48 AM
Views 2

System Information

AI Information
Framework TensorFlow Lite
Backend CPU
Device AMD Ryzen AI Max+ 395
System Information
Operating System Fedora Linux 43 (Workstation Edition)
Model Framework Desktop (AMD Ryzen AI Max 300 Series)
Motherboard Framework FRANMFCP06
CPU Information
Name AMD Ryzen AI Max+ 395
Topology 1 Processor, 16 Cores, 32 Threads
Identifier AuthenticAMD Family 26 Model 112 Stepping 0
Base Frequency 5.19 GHz
Cluster 10 Cores
Memory Information
Size 31.11 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 1623
301.9 IPS
 
Image Classification (HP)
100% 1806
335.9 IPS
 
Image Classification (Q)
99% 1839
343.0 IPS
 
Image Segmentation (SP)
100% 3093
50.1 IPS
 
Image Segmentation (HP)
100% 3137
50.9 IPS
 
Image Segmentation (Q)
98% 1607
26.1 IPS
 
Pose Estimation (SP)
100% 4422
5.16 IPS
 
Pose Estimation (HP)
100% 4441
5.18 IPS
 
Pose Estimation (Q)
96% 6261
7.34 IPS
 
Object Detection (SP)
100% 2171
172.2 IPS
 
Object Detection (HP)
100% 2138
169.6 IPS
 
Object Detection (Q)
85% 2123
170.9 IPS
 
Face Detection (SP)
100% 3677
43.7 IPS
 
Face Detection (HP)
100% 3662
43.5 IPS
 
Face Detection (Q)
97% 3600
42.9 IPS
 
Depth Estimation (SP)
100% 3132
24.1 IPS
 
Depth Estimation (HP)
99% 3184
24.5 IPS
 
Depth Estimation (Q)
63% 3015
28.3 IPS
 
Style Transfer (SP)
100% 5494
7.06 IPS
 
Style Transfer (HP)
100% 5443
7.00 IPS
 
Style Transfer (Q)
98% 9516
12.3 IPS
 
Image Super-Resolution (SP)
100% 1955
72.2 IPS
 
Image Super-Resolution (HP)
100% 1914
70.7 IPS
 
Image Super-Resolution (Q)
97% 2599
96.3 IPS
 
Text Classification (SP)
100% 1682
2.25 KIPS
 
Text Classification (HP)
100% 1558
2.08 KIPS
 
Text Classification (Q)
92% 920
1.24 KIPS
 
Machine Translation (SP)
100% 3735
64.3 IPS
 
Machine Translation (HP)
100% 3662
63.1 IPS
 
Machine Translation (Q)
58% 807
19.0 IPS
 

Compare

Set Baseline Geekbench 6 now available