Blitz Bureau
NEW DELHI: Nvidia on February 18 unveiled a series of strategic collaborations at the India AI Impact Summit, aimed at accelerating AI transformation across the country’s industrial landscape.
The partnerships range from cloud providers to global system integrators and major industrial software firms, reports Moneycontrol.com.
“AI is a five-layer cake spanning energy, shapes, infrastructure, models, and applications. We are proud to be working with India visionaries at every single layer of this stack,” said Vishal Dhupar, Nvidia’s managing director for South Asia.
Augmenting AI compute
The world’s most valuable company is teaming up with domestic operators like Yotta, L&T and E2E Networks to deliver ‘advanced AI factories’. These hubs are designed to meet India’s growing need for AI compute and enable the country to develop its own sovereign AI models and services.
The initiative is part of the IndiaAI Compute Pillar, under which the country is building out its AI cloud offerings with systems including tens of thousands of Nvidia GPUs, available to the start-up and research community at affordable costs.
Yotta is augmenting its sovereign AI Cloud platform Shakti Cloud, which is powered by over 20,000 Nvidia Blackwell Ultra GPUs.
Dhupar said that L&T is announcing a new formation to build sovereign “gigawatt AI factory” infrastructure, with a roadmap featuring a new Mumbai facility and an expansion in Chennai.
Additionally, E2E Networks is building an Nvidia Blackwell GPU cluster on its TIR cloud compute platform, hosted at the L&T Vyoma Data Center in Chennai.
Netweb Technologies is also launching its Tyrone Camarero AI Supercomputing systems built on the Nvidia Grace Blackwell architecture to expand access to Nvidia’s AI infrastructure in India.
Netweb will manufacture Nvidia GB200 NVL4 platforms under the Government’s “Make in India” mission, featuring four Nvidia Blackwell GPUs and two Nvidia Grace CPUs to power scientific computing, model training and inference.
Dhupar also highlighted AI start-ups like Sarvam, BharatGen, and Chariot that are building sovereign AI models using Nvidia Nemotron’s libraries and datasets. Meanwhile, IT firm Tech Mahindra used Nemotron to develop an 8-billion-parameter foundation model tailored for Indian languages and dialects.
Nemotron is a collection of open models, datasets, recipes, and technologies that enable startups to build efficient and specialised agentic AI systems. For the India market, Nvidia has rolled out Nemotron Personas India, a massive open synthetic data set designed to reflect the country’s diverse population and workforce. It includes 21 million fully synthetic Indic personas to enable population-scale sovereign AI development.
Organisations across the country are building AI applications with Nvidia Nemotron to support public-sector services, financial systems and enterprise operations in multiple languages, the company stated. This includes National Payments Corporation of India (NPCI), Zoho, Gnani.ai, CoRover.ai, and Tata Communications-backed Commotion.
The world’s most valuable company is teaming up with domestic operators like Yotta, L&T and E2E Networks to deliver ‘advanced AI factories’. These hubs are designed to meet India’s growing need for AI compute and enable the country to develop its own sovereign AI models and services.
Nvidia also announced partnerships with industrial software leaders Cadence, Siemens and Synopsys to help India’s largest manufacturers build ‘AI factories of the future’.
“These are software-defined facilities and our mission is to integrate AI into every single step from initial design to the final operation,” Dhupar said.
The move comes at a time when the country is looking at investing billions of dollars in new manufacturing capacities across construction, automotive and renewable energy.
This is enabling enterprises achieve rapid design iteration and operational intelligence across the energy, automotive and electronics sectors, Nvidia stated.
For example, Havells India is using a Synopsis tool to achieve six time faster simulation for their appliances, allowing them to explore more design options to optimise airflow and energy efficiencies, and achieve faster time to market.
L&T Semiconductor is using Cadence to shorten design iterations of next-generation chips. Meanwhile, Addverb is utilising Siemens and Nvidia’s Omniverse libraries along with Cosmos world foundation models to create high-fidelity digital twins allowing them to test advanced robots in simulation before they hit the floor.
TCS is deploying physical AI solutions at Tata Motors for critical tasks like quality inspection and real-time safety monitoring.
“By combining Nvidia’s full stack AI with India’s manufacturing leadership, we aren’t just witnessing the next age of industrialisation. We are building it,” Dhupar said.
IT giants like Infosys, Persistent, Tech Mahindra and Wipro are utilising Nvidia’s AI enterprise software to build and deploy AI agents across multi-billion dollar industries from finance and software development to drug discovery, telecom, and customer services.
For instance, Wipro and Nvidia partnered with a major US healthcare insurer to launch an AI-driven voice and agent assist solution for customer support. Today, AI agents handle 42 percent of the inbound calls, responding to 900 concurrent calls and 164 requests per second with sub-200 millisecond latency, Dhupar said.
Meanwhile, Infosys has developed a new small language coding model using Nvidia’s NeMo framework and integrated it within Infosys Topaz Fabric.
The 2.5 billion parameter model is trained on a curated blend of high quality code, synthetic data, mathematical reasoning and natural language inputs and supports agent development, code generation, refactoring and end-to-end software engineering workflows.
Nvidia stated that the lightweight model can be deployed across on-premises enterprise data centers, cloud environments and even standard desktops.
