Unprecedented AI compute power of 26 Tera-Operations Per Second (TOPS)
Best-in-class power efficiency with typical power consumption of 2.5W
Highest cost-efficiency (TOPS/$) compared with existing solutions
Converting AI models to run at the edge is complex, and Hailo provides a stable, well-thought-out platform with great developer support and tools
Hailo-8 is achieving overwhelmingly high performance compared to conventional products.
This processor helps us overcome all the limitations
Hailo enables us to perform more sophisticated operations with more complex networks, and thus, offer more solutions to the wide array of manufacturers we work with
With Hailo we power edge devices with renewable energy and still get more machine learning capabilities than we used to with hardwired labs
| NN Model | Input Resolution | FPS | Power (W) | FPS/W |
|---|---|---|---|---|
Classification |
||||
ResNet-50 v1 |
224×224 |
1,372 |
3.6 |
385 |
ViT_base |
224×224 |
168 |
3.1 |
53 |
Object Detection |
||||
YOLOv5m |
640×640 |
242 |
5.6 |
43 |
YOLOv8m |
640×640 |
140 |
4.7 |
30 |
YOLOv11m |
640×640 |
102 |
3.4 |
30 |
Semantic Segmentation |
||||
stdc1 |
1,024×1,920 |
58 |
3.0 |
19 |
