YOLOv8 is a versatile computer vision model from Ultralytics designed to support a broad range of vision tasks, including object detection, image classification, instance segmentation, pose estimation, and oriented object detection, with an emphasis on accuracy, speed, and ease of deployment.

Original Reference: Ultralytics YOLOv8 — Official YOLOv8 documentation

YOLOv8s

This model uses the YOLOv8-Small(YOLOv8s) variant for object detection, providing a lightweight configuration optimized for low-latency inference and small memory footprint. It is well suited for real-time object detection in applications such as automotive perception, surveillance, robotics, industrial inspection, and edge video analytics.

Model Configuration:

• Reference implementation: Ultralytics forked by Ambarella
• Original Weight: YOLOv8s_Weights.COCO2017
• Dataset: COCO
• Resolution: 3x640x640
• Support Cooper version:
  • Cooper SDK: [2.5.4]
  • Cooper Foundry: [2.3]

Model	Device	compression	Model Link
YOLOv8s	N1-655	Amba_optimized	Model_Link
YOLOv8s	N1-655	Activation_fp16	Model_Link
YOLOv8s	X7	Amba_optimized	Model_Link
YOLOv8s	X7	Activation_fp16	Model_Link
YOLOv8s	CV7	Amba_optimized	Model_Link
YOLOv8s	CV7	Activation_fp16	Model_Link
YOLOv8s	CV72	Amba_optimized	Model_Link
YOLOv8s	CV72	Activation_fp16	Model_Link
YOLOv8s	CV75	Amba_optimized	Model_Link
YOLOv8s	CV75	Activation_fp16	Model_Link