Vision Analysis
All Hardware

Raspberry Pi 5 + Hailo-8

Broadcom BCM2712

RAM

16 GB

Power (TDP)

17W

Highest Accuracy
Top 5 models by mAP@50-95
1.
yolov9c
54.8%
2.
yolov9m
53.3%
3.
yolonas-l
52.4%
4.
yolonas-m
47.4%
5.
yolonas-s
44.5%
Fastest
Top 5 models by throughput
1.
yolox-tiny
196 FPS
2.
yolox-s
96 FPS
3.
yolov9t
76 FPS
4.
yolonas-s
40 FPS
5.
yolov9s
36 FPS
All Models
10 models benchmarked on Raspberry Pi 5 + Hailo-8 (HailoRT INT8)
#ModelmAPFPSLatencyParams
1
yolov9c		54.8%21.247.1ms25.5M
2
yolov9m		53.3%30.133.2ms20.1M
3
yolonas-l		52.4%15.464.9ms67.0M
4
yolonas-m		47.4%21.746.0ms51.1M
5
yolonas-s		44.5%39.725.2ms19.0M
6
yolox-m		43.7%34.828.8ms25.3M
7
yolox-s		41.1%95.810.4ms9.0M
8
yolox-tiny		33.5%195.65.1ms5.1M
9
yolov9s		32.0%36.327.5ms7.2M
10
yolov9t		25.8%76.513.1ms2.0M

Run any model with one line

LibreYOLO has the best catalogue of state-of-the-art detectors, all behind one MIT-licensed Python API.

from libreyolo import LibreYOLO, SAMPLE_IMAGE

# LibreYOLO has the best catalogue of state-of-the-art models.
model = LibreYOLO("LibreRFDETRl.pt")           # RF-DETR-L (transformer flagship)
results = model(SAMPLE_IMAGE, save=True)        # run inference, save the annotated image

# Swap in any other model, same one-line API (weights auto-download):
#   LibreYOLO("LibreYOLO9c.pt")      # YOLO9-C
#   LibreYOLO("LibreYOLOXx.pt")      # YOLOX-X
#   LibreYOLO("LibreDFINEx.pt")      # D-FINE-X
#   LibreYOLO("LibreRTDETRr50.pt")   # RT-DETR-R50