L&T Wins Major AI Infrastructure Order as India Scales High-Performance Computing
Larsen & Toubro (L&T) has secured a major contract worth up to ₹150 billion ($1.57 billion) from U.S.-based AI cloud company Together AI to develop a large-scale AI data center in India. The project will use high-performance NVIDIA computing infrastructure and represents a significant expansion of L&T’s role in the country’s rapidly developing AI infrastructure ecosystem.
The announcement comes as demand for AI-ready data centers, high-density GPU infrastructure and accelerated computing capacity increases across India. For infrastructure companies, cloud providers, semiconductor vendors and enterprise technology buyers, the project highlights how AI is increasingly driving investment beyond software and chips into power, cooling, networking and purpose-built data center capacity.
Explore the AI Data Centers Market Report
Understand the AI infrastructure opportunity with data-driven market insights, forecasts, competitive analysis, regional trends, and emerging growth areas through 2035. Explore the full AI Data Centers Market report to support investment, expansion, and strategic planning decisions. Request a Sample Report Today!
What Is the L&T NVIDIA AI Factory Project?
The project is being developed for Together AI, an AI cloud platform focused on providing infrastructure for AI workloads. L&T's latest order is classified as a major project and is valued at up to ₹150 billion.
The facility is expected to deploy 10,000 NVIDIA B300 GPUs at L&T’s Chennai data center campus, according to reports following the announcement. The infrastructure is designed to support large-scale AI workloads, including model training, fine-tuning and inference.
This positions the development as one of the most significant single-cluster AI infrastructure projects announced in India and reinforces Chennai’s growing role in the country’s high-performance computing ecosystem.
Key Project Highlights
- Project value: Up to ₹150 billion / approximately $1.57 billion
- Developer: Larsen & Toubro
- AI cloud customer: Together AI
- GPU platform: NVIDIA B300
- Planned GPU deployment: Approximately 10,000 GPUs
- Location: L&T’s Chennai data center campus
- Primary workloads: AI training, fine-tuning and inference
- Strategic focus: High-density AI computing and scalable data center infrastructure
Why the L&T AI Factory Matters for India
The project signals a transition in India’s data center industry from conventional cloud and enterprise computing toward AI-native infrastructure.
Traditional data centers were primarily designed around comparatively predictable CPU workloads. AI infrastructure requires substantially higher computing density, faster networking, greater power availability and advanced thermal management.
Large GPU clusters can therefore change the economics and engineering requirements of data center development.
For infrastructure providers, the opportunity is expanding across several interconnected areas:
- AI-optimized data center construction
- GPU and accelerator infrastructure
- High-capacity power systems
- Liquid and advanced cooling technologies
- High-speed networking
- Data center infrastructure management
- GPU-as-a-Service and AI cloud platforms
DataM Intelligence research indicates that the global AI Data Centers Market reached approximately $120.74 billion in 2025 and is projected to reach $1.02 trillion by 2035, representing a 22.8% CAGR during 2026 - 2035. Asia-Pacific is identified as the fastest-growing regional market as India and other economies accelerate AI and cloud infrastructure investment.
NVIDIA B300 GPUs Increase Infrastructure Requirements
The planned deployment of NVIDIA B300 GPUs also highlights a broader trend: next-generation AI accelerators are increasing the infrastructure requirements of data centers.
As GPU clusters become larger and more power-dense, operators need to optimize the entire facility rather than simply increase server capacity.
This includes:
- High-density rack architecture
- Advanced power distribution
- High-speed GPU networking
- Liquid cooling and thermal management
- Reliable grid connectivity
- AI workload orchestration
- High-performance storage
- Infrastructure monitoring and optimization
DataM Intelligence research identifies high-density GPU clusters, liquid cooling and energy-efficient architectures as increasingly important differentiators for AI data center operators.
The result is a shift toward integrated AI factories, where computing, networking, power and cooling infrastructure are engineered as a single system.
L&T’s Broader NVIDIA AI Infrastructure Strategy
The latest order builds on L&T’s earlier strategy to establish large-scale NVIDIA AI infrastructure in India.
In February 2026, L&T announced a proposed venture with NVIDIA to develop gigawatt-scale AI factory infrastructure under the IndiaAI Mission. The initiative was designed to combine L&T’s engineering and infrastructure capabilities with NVIDIA’s GPUs, CPUs, networking, accelerated storage and AI software ecosystem.
L&T previously outlined plans to scale NVIDIA GPU cluster deployment at its Chennai data center and develop additional AI-ready capacity in Mumbai.
The latest Together AI order therefore represents a commercialization step for L&T’s broader AI infrastructure strategy.
Together AI Deal Strengthens India’s AI Cloud Ecosystem
The involvement of Together AI is strategically important because AI infrastructure is increasingly shifting toward specialized cloud platforms capable of providing GPU capacity to developers, enterprises and AI companies.
Instead of every organization purchasing and operating its own large GPU cluster, AI cloud providers can aggregate computing resources and offer access through cloud infrastructure.
This model can accelerate AI adoption by reducing the upfront infrastructure burden for customers while increasing utilization of expensive GPU assets.
For India, the development could strengthen the country’s position as a destination for AI computing capacity serving both domestic and international customers.
AI Infrastructure Is Becoming a Full-Stack Investment Opportunity
The L&T project demonstrates that AI infrastructure investment extends far beyond semiconductor procurement.
A large AI factory requires coordination across the entire infrastructure stack:
AI Accelerators → Servers → Networking → Storage → Power → Cooling → Data Center → Cloud Platform → AI Workloads
This creates opportunities for companies across multiple technology and industrial segments.
The strongest investment opportunities may emerge in areas such as AI data center construction, GPU infrastructure, liquid cooling, power management, data center batteries, high-speed networking and infrastructure management.
DataM Intelligence research estimates that the global AI Data Center Liquid Cooling Market could grow from $3.39 billion in 2025 to $23.23 billion by 2035, reflecting the increasing importance of thermal management as GPU density rises.
India’s AI Infrastructure Race Is Accelerating
The L&T order comes amid increasing investment in India’s digital and AI infrastructure.
India is seeking to build domestic AI capacity while attracting global technology companies, cloud providers and infrastructure investors. This is creating demand for data centers capable of supporting high-performance AI workloads within the country.
The strategic importance extends beyond computing capacity.
Domestic AI infrastructure can support:
- Data localization requirements
- Enterprise AI deployment
- Government AI applications
- Sovereign AI workloads
- Generative AI development
- Large language model training
- AI inference services
- Research and innovation
- Global AI cloud services
The combination of government-backed AI initiatives, cloud adoption and private infrastructure investment is creating a new growth cycle for India's data center industry.
Analyst View: Why This Project Could Reshape India's AI Data Center Market
The L&T–Together AI project should be viewed as more than a single data center contract. It represents the increasing commercialization of AI infrastructure in India.
The deployment of a large NVIDIA GPU cluster illustrates how AI workloads are becoming a direct driver of data center construction, power infrastructure and advanced cooling demand.
From an industry perspective, the key competitive advantage will increasingly shift from simply owning data center capacity to delivering AI-ready capacity at scale.
For infrastructure developers, this means access to power, rack density, cooling efficiency, networking performance and GPU availability will become critical factors in project economics.
For technology vendors, the opportunity is expanding into the supporting infrastructure required to operate AI factories efficiently.
For investors, the development highlights a broader value chain encompassing semiconductors, AI accelerators, servers, networking, cooling, power systems, data centers and AI cloud platforms.
What Comes Next for India's AI Factory Market?
The L&T order could encourage additional investment in AI factories and high-density data center campuses across India.
Future projects are likely to focus on:
- Larger GPU clusters
- Gigawatt-scale AI campuses
- AI-focused cloud infrastructure
- Liquid-cooled data centers
- Renewable and low-carbon power integration
- High-density AI racks
- Sovereign AI computing
- GPU-as-a-Service platforms
- Advanced networking infrastructure
The biggest constraint may increasingly shift from access to computing technology toward power availability, grid connectivity, cooling capacity and speed of infrastructure deployment.
As AI models become more computationally demanding, the ability to build and operate high-density infrastructure efficiently could become a decisive factor in national and corporate AI competitiveness.
Conclusion
L&T’s order from Together AI marks a significant milestone in India’s transition toward large-scale AI infrastructure. With a potential value of up to $1.57 billion and a planned deployment of NVIDIA-powered computing capacity, the project reinforces the growing role of Indian data center operators in the global AI infrastructure ecosystem.
The development also underscores a broader market trend: the next phase of AI investment will require not only advanced GPUs, but also purpose-built data centers, high-capacity power systems, advanced cooling, networking and cloud platforms.
As India expands its AI capacity, projects of this scale could become an important catalyst for the country's emergence as a regional hub for AI computing and digital infrastructure.