Meta and Reliance Partner for First AI Data Center in India
Strengthening the AI backbone for one of the world's fastest-growing digital markets
In a move that underscores India's critical role in the global AI landscape, Meta Platforms has teamed up with Reliance Industries to establish its first AI data center in the country. This strategic partnership aims to build a 168-megawatt facility that will serve as the foundational infrastructure for Meta's expanding suite of AI-driven services across its family of apps, including Facebook, Instagram, and WhatsApp. As the demand for real-time AI inference and localized model training skyrockets, this deal represents a significant milestone in localized compute for the subcontinent.
Key Details
The announcement, made on June 10, 2026, reveals that Reliance Industries will develop the data center in Jamnagar, Gujarat. This location is already home to the world’s largest single-site oil refinery, showcasing Reliance's ability to manage massive industrial scale and complex energy requirements. Meta will lease the 168-megawatt facility, with an agreement that includes options for significant expansion as compute demands grow over the next decade.
Meta has committed to covering the full operational costs of energy and water for the facility, aligning with its global sustainability goals and ensuring that the data center's carbon footprint is closely managed. While the exact financial terms of the lease were not disclosed, market analysts view the deal as a major strategic win for both companies. Reliance's stock responded positively to the news, gaining 2.5% in early trading as investors cheered the conglomerate's deepening involvement in the AI value chain.
This collaboration is not the first for the two giants; rather, it is the culmination of a deepening alliance. It builds upon a long-standing relationship that began with Meta's massive $5.7 billion investment in Jio Platforms in 2020. Since then, the two have worked closely on integrating WhatsApp with Jio’s commerce ecosystem. More recently, in 2025, the companies formed a joint venture specifically focused on building large-scale AI platforms and agentic tools tailored for Indian small and medium-sized enterprises (SMEs).
What This Means
For Meta, a dedicated AI data center in India is about more than just reducing latency for its hundreds of millions of users in the region. It is a strategic move to localize compute resources in a regulatory environment that is increasingly emphasizing data sovereignty and the physical location of servers. By having "AI on the ground" in India, Meta can more effectively deploy localized AI models that understand the nuances of India's 22 official languages and diverse cultural contexts, which are often lost in models trained primarily on Western datasets.
The proximity of compute also enables "edge-like" performance for complex AI features. Features like real-time voice translation in WhatsApp or sophisticated AR filters in Instagram require massive throughput and minimal lag to feel seamless. By offloading this processing to a local Jamnagar-based cluster rather than routing it through Singapore or Europe, Meta can offer a significantly superior user experience that its competitors might struggle to match.
For Reliance, the deal cements its position as the premier infrastructure provider for the digital age in India. Mukesh Ambani’s conglomerate is pivoting hard toward deep tech, and hosting a global AI leader like Meta provides a high-profile proof of concept for further large-scale data center developments. It also serves the Indian government's broader vision of "AI for All," making the country a global hub for AI research and deployment capable of competing head-to-head with the technological hegemonies of the United States and China.
Technical Breakdown
The Jamnagar facility is expected to house the latest generation of custom-designed AI accelerators and high-bandwidth memory systems, designed to handle the massive throughput required for both generative AI training and low-latency inference.
- 168 MW Capacity: To put this in perspective, this is enough power to support hundreds of thousands of high-end GPUs—such as NVIDIA's Blackwell architecture or Meta's own MTIA chips—working in parallel.
- Lease and Expand Model: This flexible arrangement allows Meta to scale its footprint dynamically without the immediate multi-billion dollar capital expenditure of building the shell and power infrastructure from scratch. Reliance provides the "dirt, power, and security," while Meta provides the silicon and software stack.
- Sustainable Operations: Gujarat is a hub for renewable energy, particularly solar and wind. Meta’s commitment to energy costs suggests a high-efficiency cooling system, possibly using liquid cooling techniques that are becoming standard for high-density AI clusters.
- Fiber Integration: The site will benefit from Jio's extensive national fiber network, ensuring massive "backhaul" capacity and ultra-high-speed data transfer between the Jamnagar cluster and Meta's global network edge nodes.
Industry Impact
This deal signals the start of a massive infrastructure "build-out" phase in the Indian market. We are seeing a historic wave of investment from major Western tech firms. Amazon and Microsoft have already committed a combined $52 billion to India this year alone, much of it earmarked for cloud and AI regions. This Meta-Reliance deal is the first major example of a "social-first" company making a heavy physical infrastructure bet in the region.
The entrance of these players into the Indian infrastructure market will likely lead to an intense "talent war" for AI engineers, data center architects, and power management specialists. Furthermore, the presence of such massive localized compute resources provides Indian startups with a psychological and practical boost. As these facilities come online, they often lead to secondary markets for specialized AI services, potentially lowering the barrier to entry for local AI innovation and "vibe coding" startups.
Looking Ahead
As the Jamnagar facility comes online and begins processing its first tokens, expect Meta to roll out more sophisticated agentic AI features specifically for its Indian user base. We may see WhatsApp become a much more powerful platform for end-to-end commerce, where AI agents handle everything from product discovery to payment and delivery tracking, all through a natural language interface that understands regional dialects.
The broader trend is undeniable: the AI race is shifting from a battle of who has the most elegant algorithms to a battle of who has the most robust physical infrastructure and the most reliable power supply. In India, the alliance between Meta and Reliance suggests that the biggest winners in the Intelligence Age will be those who can marry global software and chip prowess with massive local industrial scale and political alignment. Keep an eye on Gujarat; it’s fast becoming the silicon heart of the East.
Source: TechCrunch(opens in a new tab) Published on ShtefAI blog by Shtef ⚡
