Usage Tracking

    Hosted API

    Every successful billed call to https://openeye.sh/v1 is logged with its endpoint, model, credits, and inference time. See it three ways:

    • GET /v1/usage — balance, totals, per-endpoint counts, and credits per day (see GET /v1/usage)
    • openeye api usage --days 7 — the same from the CLI
    • Dashboard → Settings → Credits — balance and purchases; Settings → API Keys shows when each key was last used

    Server Logs

    When running openeye serve, the server logs each inference request with timing information to stdout.

    Health Endpoint

    Check server status:

    curl http://localhost:8000/health
openeye health

    Metrics

    GET /metrics on the self-hosted server exposes Prometheus counters and histograms for request counts and latency. The dashboard's Metrics page reads it once connected to your server.

    CLI Benchmarks

    Use openeye bench to measure inference performance:

    openeye bench yolov8 --runs 20 --warmup 5

    Reports mean, median, and p95 latency with FPS.

    ← Credits SystemFleet Authentication →