Geekbench ML Score

922
TensorFlow Lite CPU Inference Score
Geekbench ML 0.6.0 for Linux AVX2

Result Information

Upload Date August 16 2024 06:38 PM
Views 1

System Information

System Information
Operating System Debian GNU/Linux 12 (bookworm)
Model iXsystems RS520-E9-RS12-E-IXN
Motherboard ASUSTeK COMPUTER INC. Z11PR-D16 Series
CPU Information
Name Intel Xeon Gold 6226R
Topology 2 Processors, 32 Cores, 64 Threads
Identifier GenuineIntel Family 6 Model 85 Stepping 7
Base Frequency 3.90 GHz
Cluster 10 Cores
L1 Instruction Cache32.0 KB x 16
L1 Data Cache32.0 KB x 16
L2 Cache1.00 MB x 16
L3 Cache22.0 MB x 1
Memory Information
Size 251.51 GB
Inference Information
Framework TensorFlow Lite
Backend CPU
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 1227
229.5 IPS
 
Image Classification (F16)
100% 1247
233.4 IPS
 
Image Classification (I8)
99% 805
150.6 IPS
 
Image Segmentation (F32)
100% 1361
22.7 IPS
 
Image Segmentation (F16)
100% 1273
21.3 IPS
 
Image Segmentation (I8)
98% 689
11.5 IPS
 
Pose Estimation (F32)
100% 1549
1.88 IPS
 
Pose Estimation (F16)
100% 1679
2.03 IPS
 
Pose Estimation (I8)
100% 2476
3.00 IPS
 
Object Detection (F32)
100% 820
61.2 IPS
 
Object Detection (F16)
100% 907
67.7 IPS
 
Object Detection (I8)
65% 717
53.5 IPS
 
Face Detection (F32)
100% 1323
15.7 IPS
 
Face Detection (F16)
100% 608
7.23 IPS
 
Face Detection (I8)
87% 591
7.03 IPS
 
Depth Estimation (F32)
100% 944
7.32 IPS
 
Depth Estimation (F16)
100% 937
7.27 IPS
 
Depth Estimation (I8)
95% 1250
9.69 IPS
 
Style Transfer (F32)
100% 1795
2.36 IPS
 
Style Transfer (F16)
100% 1771
2.33 IPS
 
Style Transfer (I8)
98% 3179
4.18 IPS
 
Image Super-Resolution (F32)
100% 686
24.5 IPS
 
Image Super-Resolution (F16)
100% 1399
50.0 IPS
 
Image Super-Resolution (I8)
98% 955
34.1 IPS
 
Text Classification (F32)
100% 319
458.2 IPS
 
Text Classification (F16)
100% 548
787.9 IPS
 
Text Classification (I8)
92% 241
346.3 IPS
 
Machine Translation (F32)
100% 189
3.47 IPS
 
Machine Translation (F16)
100% 946
17.4 IPS
 
Machine Translation (I8)
62% 480
8.84 IPS
 

Compare

Set Baseline Geekbench 6 now available