What If You Never Needed an API Key Again? Building a Mesh LLM From Spare Compute

The current AI stack has a dependency most of us don't talk about: a handful of closed models from a handful of providers, and an API call standing between every agent and every action. Mic Neale — who helped build Goose, Block's open-source agentic coding system — thinks that's a problem worth solving at the infrastructure level. This talk introduces a working prototype of a decentralised mesh LLM: a system where individuals and small teams pool their spare compute to collectively run open models that none of them could run alone. When your GPU is idle, it contributes to the mesh. When you need capacity, the mesh is there. The result is access to capable open models without a cloud bill or an API dependency.

Mic will walk through how the mesh works technically — coordination, model sharding, latency management, and what happens when nodes drop out — and where it's headed: a model where contributors earn from their idle capacity, and where the economics of running frontier-class open models shift from "data centre required" to "your neighbourhood has enough."