When building software, inference is now a must-have, but getting it affordably and without self-hosting is difficult, but getting inference with datacenter level failover whilst still being affordable is much harder. There is a huge volume of under-utilized compute all around us, individuals, companies, and cloud providers, all with spare compute and a need to earn. We built a platform that makes use of compute from almost any machine around the world, and compensates the provider with payments from paying developer customers who need that reliable open model inference but at a more affordable rate. We've built around the philosophy of keeping inference affordable but reliable, and we want you to come on board as a developer or provider, and start building with powerful open models!

Scalattice

OpenAI-compatible inference on a distributed GPU network

OpenAI-compatible inference on a distributed GPU network