Equinix, NVIDIA, and Together AI have formed a strategic alliance to accelerate enterprise AI inference deployment.

Equinix, Inc. announced a significant expansion of its longstanding collaboration with NVIDIA Corporation to deliver Equinix Inference Exchange, a distributed AI inference program for global enterprises. This initiative also introduces a new partnership with Together AI, combining NVIDIA Corporation’s validated Enterprise Reference Architectures with Together AI’s inference platform to support more than 200 open-source models. Hosted across Equinix, Inc.’s global data centers, the solution provides seamless connectivity to clouds, networks, and AI providers via Equinix Fabric.

The offering will debut at Equinix Horizon, the company’s inaugural customer and partner event, alongside Equinix Fabric One, designed to simplify enterprise connections across globally distributed AI environments. Equinix Inference Exchange accelerates the transition from AI experimentation to production by delivering secure, low-latency connectivity to the data, users, and ecosystem organizations rely on. The architecture rests on three complementary layers engineered to streamline distributed AI inference.

Equinix, Inc. supplies the foundational infrastructure, including power, advanced cooling, and day-two operations, linked through Equinix Fabric to the clouds, networks, and AI providers essential for inference workloads. NVIDIA Corporation anchors the build with purpose-built AI infrastructure and Enterprise Reference Architectures optimized to maximize AI factory throughput and minimize token cost. Operating atop this foundation, Together AI delivers a flexible platform that supports multitenant deployments for shared efficiency and dedicated single-tenant environments for capacity-intensive workloads.

Built on Equinix Fabric, the integrated solution connects to inference providers across major metropolitan areas worldwide, significantly reducing time-to-first-token while simplifying deployment across an expansive ecosystem of clouds, networks, and AI providers. It is designed to accommodate a broad spectrum of enterprise inference scenarios, including metro edge inference, open model migration, and sovereign AI initiatives. Equinix Inference Exchange will be available starting in the First Quarter 2027.