The Equinix company has announced a significant expansion of its long-term collaboration with NVIDIA to launch Equinix Inference Exchange, a distributed AI inference program for global enterprises, along with a new collaboration with Together AI.
As AI expands across diverse models, vendors, and geographies, where inference is executed becomes a strategic factor that determines performance, costs, and governance. Equinix Inference Exchange will offer enterprises a faster path from AI experimentation to production, providing secure, low-latency connectivity to the data, users and ecosystem they depend on.
This collaboration unites NVIDIA’s validated enterprise reference architecture with Together AI’s inference platform, which supports more than 200 open source models. The service, which will be offered through Equinix’s global data centers, will provide connectivity to clouds, networks and AI providers using Equinix Fabric.
Connect distributed AI environments
The solution was launched at Equinix Horizon, the company’s inaugural event for customers and partners, along with Equinix Fabric One, a tool that will make it easier for companies to connect between globally distributed AI environments.
“AI is transforming enterprise technology at extraordinary speed, and the infrastructure decisions businesses make today will shape their competitive position for years to come. Equinix is uniquely positioned to deliver what this moment demands, thanks to our nearly three decades of experience building the trusted exchange point where businesses around the world run, connect and orchestrate their most critical workloads,” said Adaire Fox-Martin, CEO and president of Equinix. “Our long-standing relationship with NVIDIA provides the foundation for “Accelerated computing is at the core of modern AI, while Together AI’s commitment to open ecosystems gives businesses the flexibility to scale on their own terms. Equinix Inference Exchange will enable architectures that are neutral by design, open by default, and optimized for exceptional performance.”
“Equinix Inference Exchange transforms the world’s leading digital networking platform into a global infrastructure for AI inference,” said Raj Mirpuri, vice president of global AI clouds and infrastructure ecosystem at NVIDIA. “As accelerated computing becomes a strategic asset, the combination of NVIDIA’s infrastructure and technology with Together AI’s open model inference platform and Equinix’s global reach gives enterprises a powerful, distributed foundation to bring intelligence closer to their data, applications and customers, accelerating the next generation of intelligent services.”
“Together AI was born from the belief that open and accessible AI is what will define the future of the industry, as companies should not have to choose between model performance and operational flexibility,” said Vipul Ved Prakash, co-founder and CEO of Together AI. “What we are building with Equinix and NVIDIA demonstrates that model choice and performance are not mutually exclusive; They form the basis of well-implemented enterprise AI.”
Where inference is executed is key
The pace of adoption of enterprise AI is outpacing the capacity of the infrastructure needed to support it. As enterprise AI moves from experimentation to production, inference needs to run increasingly closer to the users, data, and applications it serves, spanning multiple clouds, models, vendors, and geographies. This change forces companies to determine not only how to deploy AI infrastructure, but also where it should run and how it connects to the data, applications and workloads on which it depends.
Managing these distributed inference deployments involves considerable operational complexity, precisely at a time when enterprises need greater control and visibility.
“Performance, cost and governance have become strategic drivers as AI workloads become more distributed across vendors, data sources and environments,” said Nick Patience, vice president and head of the AI Platforms practice at The Futurum Group. “Organizations are increasingly focused on where inference runs and how quickly it can be deployed in production. Solutions that simplify inference deployment without sacrificing flexibility will become increasingly important to achieving business objectives.”
Equinix brings unmatched scale and ecosystem density to address diverse challenges
Equinix brings unmatched scale and ecosystem density to meet this challenge, with more than 280 data centers in 77 metropolitan areas, 230 cloud access points and more than 10,500 companies interconnected on its neutral exchange platform. Eight of the top 10 AI model vendors and nine of the top 10 AI clouds operate with Equinix, underscoring the company’s position at the center of the AI ecosystem.
Designed for choice and flexibility
Together AI is the latest addition to Equinix’s broad AI ecosystem, bringing flexibility and choice—through open models—to companies deploying AI at scale. The solution combines three complementary layers designed to simplify distributed AI inference:
- Equinix provides the base infrastructure—including power, advanced cooling, and ongoing maintenance and management operations—connected through Equinix Fabric to the clouds, networks, and AI providers on which inference depends.
- NVIDIA supports the solution with its enterprise reference architectures and AI infrastructure designed specifically to maximize AI Factory performance and minimize cost per token.
- Together AI operates the underlying platform, supporting both multi-tenant deployments for shared efficiency and dedicated single-tenant environments for workloads that require dedicated capacity.
Based on Equinix Fabric, the solution connects with inference providers in major metropolitan areas around the world, reducing time to first token. Additionally, it will connect to a broad ecosystem of clouds, networks and AI providers, simplifying implementation complexity.
Designed for modern business inference
The solution aims to support a wide range of business inference scenarios, including:
- Metro edge inference: For organizations that need to run inference closer to users and data, enabling AI experiences with lower latency while leveraging the security, operational scale, and global reach of Equinix.
- Migration to open models: For companies that move workloads from closed and proprietary models to open source alternatives in order to control costs and avoid dependence on a single supplier, the solution will offer a direct and simple path to execute said migration in production; Together AI’s open model platform will be accessible through the same interconnected infrastructure that companies already use to connect with other providers.
- Sovereign AI: For companies operating in regulated sectors or specific geographic regions, the solution will enable AI workloads to run in locations that meet data residency and sovereignty requirements, facilitating the deployment of AI at scale while maintaining control over where data and inference are processed.
Equinix Inference Exchange will be available starting in the first quarter of 2027.
