Breaking News
Indian artificial intelligence startup Sarvam has secured $234 million in the first close of its Series B funding round, valuing the company at $1.5 billion as it ramps up investments in frontier AI models and sovereign AI infrastructure.
The round, part of a planned $300 million raise, is led by HCLTech with a $150 million strategic investment and includes participation from Bessemer Venture Partners, while existing backers Khosla Ventures and Peak XV Partners also doubled down on their support.
The fresh capital will be used to advance Sarvam's research on next-generation AI models focused on agentic AI, coding assistants and cybersecurity applications, while expanding access to large-scale computing infrastructure and accelerating deployments across enterprise and government sectors.
Founded with the ambition of building a full-stack AI ecosystem in India, Sarvam develops its own foundation models, inference infrastructure and enterprise AI platforms, positioning itself as a sovereign alternative for organizations seeking greater control over AI deployments and data.
"Our focus is on research-led innovation that delivers AI built for India's scale," said Pratyush Kumar, co-founder of Sarvam. "That means models that understand Indian languages, process local documents and provide intelligence at a cost that enterprises and governments can adopt widely."
The investment also deepens the strategic partnership between Sarvam and HCLTech, which plans to combine its enterprise transformation expertise, engineering capabilities and global customer relationships with Sarvam's AI research to build secure and scalable AI solutions.
"Our investment in Sarvam marks a significant step toward building India's trusted and globally competitive AI ecosystem," said C Vijayakumar, CEO and Managing Director of HCLTech. "Together, we are creating a differentiated full-stack AI platform for enterprises and governments."
Sarvam has rapidly expanded its AI portfolio over the past few months. Its flagship 105-billion-parameter model is designed for reasoning and agentic AI workloads, while its 30-billion-parameter model targets edge deployments capable of running on consumer hardware.
The company has also developed specialized vision and speech models for Indian use cases. Its document AI platform has digitized more than 35 million pages, including insurance records and land documents, while its speech models transcribe over 500,000 hours of audio every month across multiple Indian languages.
Beyond model development, Sarvam is seeing growing adoption of its enterprise products. Its conversational AI platform now handles more than 2 million interactions daily, with usage doubling over the past two months. The company's inference platform processes around 10 million API calls each day, with demand tripling in the last three months.
Sarvam's agentic AI platform is also gaining traction in financial services, where a leading fintech company is using it to support a sales network of approximately 350,000 representatives.
The startup is increasingly deploying AI at population scale through government and enterprise projects. Its multilingual voice agents recently collected data from 17 million farmers for the Ministry of Agriculture and Farmers' Welfare, while a nationwide voice campaign for a major insurance provider supported policy renewals for 45 million customers.
"Our ambition is to make AI widely accessible across India and create value for citizens, businesses and governments," said co-founder Vivek Raghavan. "The partnership with HCLTech demonstrates how Indian enterprises can help build foundational AI capabilities for the country."
Investors view the company as one of India's strongest bets in sovereign AI. "Sarvam is building and deploying India's sovereign AI platform for 1.4 billion citizens, mission-critical sectors and large enterprises," said Pankaj Mitra, Partner at Bessemer Venture Partners.
The funding comes as India intensifies efforts to develop indigenous AI capabilities and reduce reliance on overseas foundation models. With enterprises and government agencies increasingly prioritizing data sovereignty, multilingual AI and localized infrastructure, Sarvam is positioning itself as a homegrown alternative capable of serving regulated sectors such as banking, insurance, defence and public administration.
The round, part of a planned $300 million raise, is led by HCLTech with a $150 million strategic investment and includes participation from Bessemer Venture Partners, while existing backers Khosla Ventures and Peak XV Partners also doubled down on their support.
The fresh capital will be used to advance Sarvam's research on next-generation AI models focused on agentic AI, coding assistants and cybersecurity applications, while expanding access to large-scale computing infrastructure and accelerating deployments across enterprise and government sectors.
Founded with the ambition of building a full-stack AI ecosystem in India, Sarvam develops its own foundation models, inference infrastructure and enterprise AI platforms, positioning itself as a sovereign alternative for organizations seeking greater control over AI deployments and data.
"Our focus is on research-led innovation that delivers AI built for India's scale," said Pratyush Kumar, co-founder of Sarvam. "That means models that understand Indian languages, process local documents and provide intelligence at a cost that enterprises and governments can adopt widely."
The investment also deepens the strategic partnership between Sarvam and HCLTech, which plans to combine its enterprise transformation expertise, engineering capabilities and global customer relationships with Sarvam's AI research to build secure and scalable AI solutions.
"Our investment in Sarvam marks a significant step toward building India's trusted and globally competitive AI ecosystem," said C Vijayakumar, CEO and Managing Director of HCLTech. "Together, we are creating a differentiated full-stack AI platform for enterprises and governments."
Sarvam has rapidly expanded its AI portfolio over the past few months. Its flagship 105-billion-parameter model is designed for reasoning and agentic AI workloads, while its 30-billion-parameter model targets edge deployments capable of running on consumer hardware.
The company has also developed specialized vision and speech models for Indian use cases. Its document AI platform has digitized more than 35 million pages, including insurance records and land documents, while its speech models transcribe over 500,000 hours of audio every month across multiple Indian languages.
Beyond model development, Sarvam is seeing growing adoption of its enterprise products. Its conversational AI platform now handles more than 2 million interactions daily, with usage doubling over the past two months. The company's inference platform processes around 10 million API calls each day, with demand tripling in the last three months.
Sarvam's agentic AI platform is also gaining traction in financial services, where a leading fintech company is using it to support a sales network of approximately 350,000 representatives.
The startup is increasingly deploying AI at population scale through government and enterprise projects. Its multilingual voice agents recently collected data from 17 million farmers for the Ministry of Agriculture and Farmers' Welfare, while a nationwide voice campaign for a major insurance provider supported policy renewals for 45 million customers.
"Our ambition is to make AI widely accessible across India and create value for citizens, businesses and governments," said co-founder Vivek Raghavan. "The partnership with HCLTech demonstrates how Indian enterprises can help build foundational AI capabilities for the country."
Investors view the company as one of India's strongest bets in sovereign AI. "Sarvam is building and deploying India's sovereign AI platform for 1.4 billion citizens, mission-critical sectors and large enterprises," said Pankaj Mitra, Partner at Bessemer Venture Partners.
The funding comes as India intensifies efforts to develop indigenous AI capabilities and reduce reliance on overseas foundation models. With enterprises and government agencies increasingly prioritizing data sovereignty, multilingual AI and localized infrastructure, Sarvam is positioning itself as a homegrown alternative capable of serving regulated sectors such as banking, insurance, defence and public administration.
See What’s Next in Tech With the Fast Forward Newsletter
Tweets From @varindiamag
Nothing to see here - yet
When they Tweet, their Tweets will show up here.
