Post

federated agent marketplace - gpu/llm rental with machine-to-machine payments workers broadcast agent workload capabilities (gpu, models, cost) via @iroh.computer gossip agents select a worker based on needs (do X on qwen3-coder-480b) parties transact via Stripe MPP/HTTP 402 is this a good idea?

I suspect you'll want *some* kind of centralized system doing the brokerage instead of a permissionless iroh-gossip swarm. That said, using iroh itself for streaming tokens between workers and consumers is a great idea!

Haven’t tested the x402 in a while started focusing on free but I think this spec can do what you’re looking for

Those are exactly why I haven’t tested the x402 in a while and was focusing ok market.bids.free! :P ”No one can serve two masters. Either you will hate the one and love the other, or you will be devoted to the one and despise the other. You cannot serve both God and money.“

*on I believe it’s important for us to leverage technology to help us do and automate doing good things for one another. Sharing resources is on of those things, compute happens to be an easy target since the whole flow can be done digitally

@pdxhackerspace.org was also thinking about sharing LLM interference endpoints with their users. @cocore.dev is best for open public network but this market spec can help with sharing within or across communities leveraging the existing social graph