Equinix, Nvidia and Together AI Launch Inference Exchange for Enterprise Open-Model Serving

Equinix unveiled Inference Exchange, a distributed enterprise-inference platform combining Nvidia Enterprise Reference Architectures, Together AI's open-model inference (200+ models) and Equinix's 280+ data centers, reachable via Equinix Fabric. Aimed at moving AI workloads out of centralized clouds and closer to enterprise data and users; general availability expected Q1 2027. Equinix shares rose 2% on the news; eight of the top 10 AI model providers and nine of the top 10 AI clouds already sit on Equinix infrastructure.