Ready Server is launching a preview of a OpenAI-compatible, shared-GPU inference service that lets customers run AI workloads locally.