your Mac is unemployed.

powerful compute. sitting idle.

we’re giving it a job.

A fleet of Macs. One endpoint.

Silicon Network runs large open models on a fleet of Apple Silicon, behind one endpoint that speaks the OpenAI, Claude, and Ollama wire formats. If you already call an OpenAI-compatible API, you are a base URL away.

You pay per token for the model you actually call, from prepaid credit. No seats, no subscription, no idle capacity to rent.

You can run a node on your own hardware too. A local node is the whole data path.

Two links. read the docs for the quickstart, and create an account for a key.

follow along on X