{% extends "_base.html" %} {% block content %}
Distributed · inference

Cluster

Run a quant too big for one Mac across every Mac on your Thunderbolt ring. Peers are discovered automatically — no IPs, no hostfiles. A model's layers split across the machines by RAM and run resident, so a 46 GB quant that SSD-streams on one Mac runs in memory across two.

nodes
combined RAM
thunderbolt
status
Topology

Serve a model on the cluster

Launches pipeline-parallel serving with an OpenAI endpoint. Refuses a slow link by default (pipeline over Wi-Fi is slower than one Mac) — pass a fast Thunderbolt ring and it flies.

This cluster is your active inference backend
Chat and Arena now run across the ring — nothing else to switch.
Open Chat →
{% endblock %}