# Talos worker

Talos worker is a downloadable GPU client that pairs with a Talos network account and serves open-model inference jobs from that account's local Ollama installation, reporting uptime toward a share of usage revenue.

Talos worker is the client component of the Talos network, a network for sharing spare GPU capacity to serve open-model inference jobs. After installing it locally, the machine is paired to a Talos account with a one-time device code, and then communicates with the Talos web app only over a WebSocket, exchanging jobs and heartbeats.

Once running, it serves inference requests from the network using a local Ollama installation with at least one pulled model, defaulting to an NVIDIA GPU when one is available and falling back to CPU otherwise. A local dashboard shows live status and a slider for how much of the machine's capacity to allocate to the network.

It reports uptime while connected, and earnings are credited per job served and shown on the Talos dashboard.

Website: https://github.com/jmerelnyc/Talos