CoreWeave Targets India’s AI Compute Market

CoreWeave is entering India with a planned 240 MW AI data center deployment at AdaniConneX’s Taloja campus in Navi Mumbai. The project will comprise three 80 MW buildings, with CoreWeave operating as the sole tenant across the initial deployment. The company expects the first phase to become available in mid-2028, with further capacity coming online in stages. CoreWeave will also have an option to add another 240 MW, potentially taking its total footprint at the campus to 480 MW.

The India expansion gives CoreWeave a substantial physical presence in one of the world’s fastest-growing AI markets. The company plans to deploy NVIDIA Vera Rubin infrastructure across the campus for training, inference, reasoning and agentic workloads. Those workloads require more than accelerator availability because power density, networking and thermal management increasingly determine usable compute. The project therefore places physical infrastructure at the center of CoreWeave’s strategy for serving Indian AI customers.

Sachin Jain, Chief Operating Officer of CoreWeave, said: “India has the talent, developers, and ambition to become a major center of AI innovation,” said Sachin Jain, Chief Operating Officer of CoreWeave. “We are investing in the infrastructure, technology, and local presence needed to support that growth, bringing 240 megawatts of capacity, NVIDIA Vera Rubin, and CoreWeave’s expertise to customers building and scaling AI in India.” His comments frame the project as a broader market investment rather than simply another regional capacity addition.

For Indian customers, the location of that compute could become an important consideration as AI workloads expand. Local infrastructure can influence latency, workload placement, data-handling requirements and the economics of running large AI applications. It can also reduce reliance on capacity located outside the country for certain workloads. CoreWeave is therefore positioning the Taloja deployment around the broader requirements of AI infrastructure, rather than compute capacity alone.

Taloja Campus Built Around High-Density AI

The planned Taloja facilities reflect the changing physical requirements of large-scale AI computing. CoreWeave says the GPU data halls will use direct liquid cooling and a non-evaporative chilled water system. That design addresses the thermal demands created by increasingly dense accelerator deployments inside modern AI clusters. Cooling capacity can become a direct constraint on usable compute when facilities operate at much higher power densities.

The three-building configuration also creates a sizeable dedicated environment for CoreWeave’s AI cloud infrastructure. Each building will provide 80 MW of capacity, creating large blocks of infrastructure within the same campus. CoreWeave’s sole-tenant position gives it greater control over infrastructure configuration and future deployment decisions. That control becomes valuable when accelerator generations, rack densities and cooling requirements change during a facility’s operating life.

The decision to design around direct liquid cooling also points to the changing economics of AI data center development. Conventional cooling architectures can face greater challenges as accelerator power levels and rack densities continue to rise. Liquid cooling allows operators to address heat closer to the source, which can support higher-density computing environments. CoreWeave’s approach suggests that thermal infrastructure will form part of the initial compute strategy rather than become a retrofit requirement later.

The campus is also designed to accommodate future generations of AI computing. That flexibility matters because accelerator technology can change faster than the physical infrastructure supporting it. A facility that cannot accommodate higher densities may require expensive modifications before its power capacity reaches its practical limit. CoreWeave’s design therefore gives the company greater scope to adapt the campus as AI hardware evolves.

India Expansion Adds Local Operating Presence

CoreWeave plans to establish an office in India and hire locally as part of the expansion. The move gives the company a direct operating presence for customer engagement, partnerships and infrastructure support. It also indicates that CoreWeave views India as a strategic AI market rather than simply another destination for data center capacity. A local team can help connect infrastructure availability with the requirements of customers building and deploying AI applications.

India’s AI ecosystem is developing across enterprises, startups, research organizations and technology companies. Those organizations increasingly need access to specialized infrastructure for model development, training and inference. CoreWeave’s dedicated AI cloud model gives it an opportunity to serve that demand with infrastructure designed specifically around accelerator-heavy workloads. The success of the strategy will depend on how effectively the company converts physical capacity into accessible and reliable AI services.

The deployment also comes as Indian data center development increasingly incorporates AI-specific infrastructure requirements. Power availability, cooling architecture and expansion capacity now influence how operators evaluate major AI facilities. A 240 MW project designed around high-density computing can therefore have a different infrastructure profile from a conventional enterprise data center. CoreWeave’s Taloja campus reflects that shift by integrating specialized cooling and next-generation accelerator infrastructure from the outset.

The local presence could also become important as CoreWeave builds relationships with Indian customers. AI infrastructure decisions often involve longer planning cycles because customers must align compute requirements with applications, budgets and deployment schedules. Having teams on the ground can help CoreWeave respond to those requirements more directly. It can also support the development of partnerships around an increasingly competitive Indian AI infrastructure market.

CoreWeave Extends Its Asia-Pacific Footprint

India expands CoreWeave’s geographic strategy across the Asia-Pacific region following its expansion into Indonesia. The company said it had approximately 4.2 gigawatts of contracted power as of its Aug. 11, 2026 earnings call. That figure illustrates the scale of infrastructure commitments supporting CoreWeave’s broader expansion strategy. The India deployment adds another major market to an infrastructure footprint spanning North America, Europe and Asia-Pacific.

Geographic expansion gives AI cloud providers more options for positioning workloads closer to customers. It can also help customers manage latency, data requirements and regional capacity needs across different applications. For CoreWeave, a broader infrastructure footprint creates more opportunities to serve customers that require AI compute in specific markets. The strategy also reduces the need to treat every AI workload as a candidate for a small number of centralized computing regions.

The India deployment will not provide immediate capacity because the first phase is expected in mid-2028. That timeline leaves a substantial execution period for power delivery, construction, cooling installation and accelerator deployment. Commissioning will also determine how quickly the planned infrastructure becomes usable capacity for customers. The distinction matters because announced megawatts do not automatically translate into operational AI compute.

The project therefore adds another test of CoreWeave’s ability to execute large infrastructure deployments at scale. AI customers increasingly care about when capacity becomes usable, not simply when a data center receives an investment announcement. Power readiness, cooling performance and cluster commissioning will determine the practical value of the Taloja campus. CoreWeave’s expansion strategy will ultimately be measured by that operational delivery.

The Strategic Importance of 240 MW

The initial 240 MW deployment gives CoreWeave a significant starting position in India while preserving an additional expansion path. The option for another 240 MW could eventually double the campus footprint without requiring the company to establish an entirely separate location. That flexibility can become important if customer demand grows faster than the initial deployment can accommodate. It also gives CoreWeave a clearer mechanism for scaling capacity around an existing infrastructure platform.

For customers, the expansion option could provide greater confidence around future capacity planning. Large AI deployments rarely remain static because model sizes, inference volumes and application requirements continue to change. Customers may therefore need access to additional compute without redesigning their infrastructure strategy around an entirely new region. A scalable campus can provide a more direct path for that growth when power and cooling infrastructure can expand alongside compute.

The larger question is how quickly India’s market can absorb specialized AI infrastructure at this scale. Demand is developing across companies building models, deploying generative AI and integrating inference into commercial applications. CoreWeave’s model focuses on providing specialized compute for customers that need to scale those workloads. The Taloja campus gives the company a large physical platform from which to compete for that demand.

The project also highlights why AI data center capacity needs to be evaluated beyond headline megawatts. Customers ultimately need to understand accelerator availability, networking performance, cooling capability and the reliability of the supporting power infrastructure. A 240 MW facility can provide substantial infrastructure scale, but its commercial value depends on how much dependable AI compute that power can support. CoreWeave’s focus on liquid cooling and next-generation AI infrastructure makes that distinction particularly relevant.

India Becomes a Strategic AI Infrastructure Market

CoreWeave’s Taloja project places India firmly within its long-term AI infrastructure expansion strategy. The combination of 240 MW, dedicated buildings, liquid cooling and NVIDIA Vera Rubin infrastructure gives the company a substantial platform for the market. The additional 240 MW option provides another layer of potential growth if demand supports further expansion. The local office and hiring plans also indicate that CoreWeave intends to build a lasting operating presence around the infrastructure.

The project shows how AI cloud providers are changing the way they approach regional expansion. Traditional cloud deployments can prioritize broad geographic coverage and general-purpose infrastructure. AI infrastructure requires a tighter relationship between accelerator architecture, power density, thermal management and network design. CoreWeave’s India strategy reflects that more specialized model.

For Indian customers, the arrival of another large AI infrastructure provider could expand the pool of specialized compute available within the country. That could influence where companies place training workloads, how they approach inference deployments and how they plan future AI capacity. It may also increase competition among infrastructure providers seeking to serve India’s growing AI ecosystem. The resulting market could place greater emphasis on usable compute, infrastructure reliability and expansion flexibility.

CoreWeave’s entry ultimately represents more than a new data center announcement. It brings a large AI-focused infrastructure platform into India at a time when power, cooling and compute availability are becoming closely connected. The Taloja campus gives CoreWeave room to build a major regional position while retaining the option to expand significantly. If the company delivers the planned capacity on schedule, the project could become an important part of India’s next phase of AI infrastructure growth.