{% extends "base.html" %} {% block main %}
Every benchmark you ran, newest first.
| Run | Started | Model | Label | Status | |
|---|---|---|---|---|---|
| #{{ r.id }} | {{ r.started_at[:16].replace("T", " ") }} | {{ r.served_model or "" }} | {% if r.id == demo_id %}demo sample run shipped with kneebench{% else %}{{ r.label or "" }}{% endif %} | {% if r.status == "running" %}{% endif %}{{ r.status }} | {% if r.status == "running" %}Follow{% else %}Report {% endif %} |
Point kneebench at an OpenAI-compatible vLLM endpoint. It works out what the hardware allows, measures what you get, and explains the gap.