three.ws · Decentralized inference
Same prompt, two providers, real responses. The left card hits the three.ws inference chain, which answers from whichever provider is fastest to respond. The right card routes the identical request through a decentralized AI gateway whose orchestrators run open-weight models on commodity GPUs. Phase 4 of the roadmap: decoupling inference from any single vendor so agents stay online when one provider doesn't.
Settings · decentralized gateway model
three.ws default
three.ws inference chainthree.ws decentralized
Open-weight · decentralizedPowered by Livepeer AI Gateway on the decentralized side. The default reply comes from the three.ws provider chain, and each card names the provider and model that actually answered it. The public decentralized gateway takes no key, so it is rate-limited and often has no orchestrator online for a given model.