Run YOLOv8 locally with OpenEye

    Real-time object detection with 80 COCO classes. Powers the safety-critical fast detection layer. Supports ONNX and TensorRT backends. OpenEye, the open-source perception CLI, downloads YOLOv8, runs it on your CPU or GPU, and returns structured JSON you can pipe into scripts, agents, and robots.

    TaskObject detection
    CLI idyolov8
    CreatorUltralytics
    Weightsultralytics/yolov8 (~6 MB)
    HardwareCPU, NVIDIA CUDA, Apple Silicon (MPS)
    Install extraopeneye-sh[yolo]

    1. Install

    pipx install "openeye-sh[yolo]"

    2. Pull YOLOv8

    openeye pull yolov8

    3. Run inference

    Output is JSON on stdout, so it composes with jq and any other Unix tool.

    openeye run yolov8 photo.jpg --pretty

    4. Serve it as an API

    Start a local server with REST and WebSocket endpoints for apps, agents, and live camera streams.

    openeye serve yolov8

    See the serve reference, REST endpoints, and Python SDK.

    Other models