~/launch80 $[discord]Join the community - talk local AI with other engineers
Documentation menu

Telemetry dashboard

L80 dashboard opens the AI command dashboard: GPU power, temperatures, VRAM and engines, live vLLM and gateway state, and per-model performance.

Free, no token

Run it

$ L80 dashboard                  # the terminal view$ L80 dashboard serve            # sampler and web page on 127.0.0.1:8787$ L80 dashboard serve --force    # start even if a sampler already answers on the port

How it behaves

  • Read-only: it watches, never changes anything.
  • One sampler per machine. Any other view attaches to it; a second sampler on the same port is refused.
  • The web page binds to the loopback address only.
  • Tuned for Intel Arc today; other GPUs get a generic view.
  • L80_DASHBOARD runs your own copy instead of the pinned one.

In L80 tui the dashboard takes over the terminal and hands it back when it exits.