Geekbench ML Score

2493
TensorFlow Lite CPU Inference Score
Geekbench ML 0.6.0 for Linux AVX2

Result Information

Upload Date August 17 2024 06:22 PM
Views 2

System Information

System Information
Operating System Ubuntu 22.04.4 LTS
Model Gigabyte Technology Co., Ltd. B650 AORUS ELITE AX
Motherboard Gigabyte Technology Co., Ltd. B650 AORUS ELITE AX
CPU Information
Name AMD Ryzen 9 7900X
Topology 1 Processor, 12 Cores, 24 Threads
Identifier AuthenticAMD Family 25 Model 97 Stepping 2
Base Frequency 5.73 GHz
Cluster 10 Cores
L1 Instruction Cache32.0 KB x 12
L1 Data Cache32.0 KB x 12
L2 Cache1.00 MB x 12
L3 Cache32.0 MB x 2
Memory Information
Size 30.49 GB
Inference Information
Framework TensorFlow Lite
Backend CPU
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 1683
314.9 IPS
 
Image Classification (F16)
100% 1675
313.5 IPS
 
Image Classification (I8)
99% 1434
268.3 IPS
 
Image Segmentation (F32)
100% 2806
46.9 IPS
 
Image Segmentation (F16)
100% 2778
46.4 IPS
 
Image Segmentation (I8)
98% 1378
23.0 IPS
 
Pose Estimation (F32)
100% 4019
4.87 IPS
 
Pose Estimation (F16)
100% 4017
4.86 IPS
 
Pose Estimation (I8)
100% 4869
5.90 IPS
 
Object Detection (F32)
100% 1814
135.4 IPS
 
Object Detection (F16)
100% 1871
139.7 IPS
 
Object Detection (I8)
65% 1766
131.8 IPS
 
Face Detection (F32)
100% 3117
37.1 IPS
 
Face Detection (F16)
100% 3102
36.9 IPS
 
Face Detection (I8)
87% 2951
35.1 IPS
 
Depth Estimation (F32)
100% 3208
24.9 IPS
 
Depth Estimation (F16)
100% 3176
24.6 IPS
 
Depth Estimation (I8)
95% 3184
24.7 IPS
 
Style Transfer (F32)
100% 4397
5.78 IPS
 
Style Transfer (F16)
100% 4435
5.83 IPS
 
Style Transfer (I8)
98% 8282
10.9 IPS
 
Image Super-Resolution (F32)
100% 1884
67.3 IPS
 
Image Super-Resolution (F16)
100% 1875
66.9 IPS
 
Image Super-Resolution (I8)
98% 2415
86.2 IPS
 
Text Classification (F32)
100% 1581
2.27 KIPS
 
Text Classification (F16)
100% 1569
2.25 KIPS
 
Text Classification (I8)
92% 836
1.20 KIPS
 
Machine Translation (F32)
100% 3717
68.4 IPS
 
Machine Translation (F16)
100% 3460
63.7 IPS
 
Machine Translation (I8)
62% 1159
21.3 IPS
 

Compare

Set Baseline Geekbench 6 now available