Benchmarks

The  Hailo-8™ AI Processor brings industry-leading neural processing throughput and power efficiency to support a wide range of AI applications.

 

The table below demonstrates some of Hailo-8’s best-in-class capabilities in the foundational neural tasks of object detection and classification. It achieves very high FPS and power efficiency (FPS/W) across a number of industry-standard neural network models, while also offloading postprocessing from the host processor its working with.

 

Benchmarks

 NN Model Type Input Resolution FPS Power [W] FPS/ W
ResNet_v1_50 Classification 224×224 1,328 3.1 428
MobileNet_v2_1.0 Classification 224×224 2,613 2.1 1267
MobileNet_v3 * Classification 224×224 3,468 1.8 1878
Regnetx-800MF Classification 224×224 2,447 2.0 1232
EfficientNet_M Classification 240×240 889 3.2 278
SSD_MobileNet_v1 Object Detection 300×300 1,038 2.3 452
Tiny_YOLOv3 Object Detection 416×416 1,010 3.2 315
YOLOv3 ** Object Detection 608×608 60 4.2 14
YOLOv4 ** Object Detection 512×512 72 3.1 23
YOLOv5m Object Detection 640×640 218 4.3 50

* MobileNet_v3 – the benchmarked model flavor is MobileNet_v3 Large Minimalistic
** Figure for processing 8 simultaneous video streams

Notes:
– Based on SDK version 3.14.0 (Q1 2022)
– Measurements are done at room temperature on a single Hailo-8 device through PCIe interface on a Hailo evaluation board
– System host: Intel® Core™ i5-9400 CPU @ 2.90GHz

Hailo-8 outperforms other edge AI processors on the market on common benchmarks.
Click here and here to see how it compares.