Run RF-DETR locally with OpenEye

    Real-time detection transformer without NMS post-processing. End-to-end architecture for high-accuracy detection. OpenEye, the open-source perception CLI, downloads RF-DETR, runs it on your CPU or GPU, and returns structured JSON you can pipe into scripts, agents, and robots.

    TaskObject detection
    CLI idrfdetr
    CreatorRoboflow
    Weightsrfdetr/rfdetr-base (~120 MB)
    HardwareCPU, NVIDIA CUDA
    Install extraopeneye-sh[rfdetr]

    1. Install

    pipx install "openeye-sh[rfdetr]"

    2. Pull RF-DETR

    openeye pull rfdetr

    3. Run inference

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

    openeye run rfdetr 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 rfdetr

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

    Other models