Sarvam AI’s $75 Million Funding Plan Could Strengthen India’s Own AI Stack

Sarvam AI is again in the funding spotlight. The Bengaluru-based artificial intelligence startup is reportedly set to raise $75 million from investors including Nvidia, Glade Brook Capital and others, according to reports.

The round matters because Sarvam is not building a normal software product. It is building India-focused AI models, speech tools, translation systems and enterprise AI infrastructure that can understand Indian languages, accents and local use cases.

The funding is also coming at a time when India is trying to build more of its own AI capability instead of depending only on foreign large language models. For Sarvam, this capital could help it train stronger models, scale its products, hire top AI talent and work with government and enterprise customers.

What Sarvam AI does

Sarvam AI builds artificial intelligence models and tools for Indian languages and Indian users.

Its products include speech-to-text, text-to-speech, translation, document intelligence and chat models. In simple words, Sarvam wants AI to work smoothly for people who speak Hindi, Tamil, Telugu, Bengali, Kannada, Marathi, Gujarati, Malayalam, Punjabi, Odia and other Indian languages.

This is important because India is not an English-only market. A banking customer may speak in Hinglish. A farmer may ask a question in Telugu. A government worker may need to process documents in Hindi or Marathi. A call centre may handle customers speaking multiple languages in one conversation.

Global AI models can handle some of this, but India needs systems tuned for local speech, scripts, accents and mixed-language use.

Founders and company background

Sarvam AI was founded in August 2023 by Dr. Vivek Raghavan and Dr. Pratyush Kumar.

Vivek Raghavan has experience in India’s digital public infrastructure ecosystem, while Pratyush Kumar has worked deeply on open-source AI for Indian languages. This background gives Sarvam a strong India-first identity.

In December 2023, Sarvam raised $41 million in Series A funding led by Lightspeed, with support from Peak XV Partners and Khosla Ventures. At that time, it was one of the largest early-stage AI funding rounds by an Indian startup.

Since then, the company has moved quickly from research ambition to actual model releases and enterprise partnerships.

Details of the new funding report

According to report, Sarvam AI is set to raise $75 million as part of its next funding move. Nvidia and Glade Brook Capital are among the investors linked to the round.

Earlier reports had also said Sarvam was in talks for a much larger funding round, with names such as Nvidia, Amazon, HCLTech, Bessemer Venture Partners, Prosperity7 Ventures, Peak XV Partners, Lightspeed and Khosla Ventures discussed around the company.

For now, the safest way to read the current update is this: Sarvam is reportedly raising $75 million, and the company has not publicly announced all final terms through its own official channels at the time of writing.

That distinction matters. Startup funding talks can change before closing.

Why Nvidia’s name matters

Nvidia’s involvement is important because AI companies need huge computing power.

Training and running large AI models requires advanced GPUs. These chips are expensive and often hard to access because global demand is very high. For a company like Sarvam, access to Nvidia’s ecosystem can help with model training, inference, optimization and enterprise deployment.

Think of it like building a large factory. AI models are the product, but GPUs are the heavy machinery. Without enough compute, even a strong AI team cannot move fast.

Nvidia has also featured Sarvam as a sovereign AI case study, highlighting the startup’s use of Nvidia’s accelerated computing platform for India-focused AI systems.

Why Glade Brook’s participation is notable

Glade Brook Capital is known for backing high-growth technology companies. Its reported interest in Sarvam shows that global investors are watching India’s AI market more seriously.

India has a large user base, many languages, strong engineering talent and growing enterprise demand for AI. But building foundation models is expensive. It needs patient investors who understand that returns may take time.

If investors like Glade Brook participate, it can improve Sarvam’s credibility in global AI circles and help the company access stronger networks.

What Sarvam may use the money for

The funding can support Sarvam in several practical ways.

1. First, model development. Sarvam has already released models such as Sarvam-30B and Sarvam-105B. The company may use fresh capital to improve reasoning, voice, translation, document understanding and Indian-language performance.
2. Second, compute. AI training and deployment need costly infrastructure. More capital can help Sarvam secure compute capacity and optimise its models for real-world use.
3. Third, enterprise growth. Sarvam is targeting businesses and government use cases. This needs sales teams, customer support, security work and industry-specific solutions.
4. Fourth, talent hiring. AI researchers, infrastructure engineers, product builders and deployment experts are expensive and in high demand.
5. Fifth, sovereign AI projects. Sarvam has been selected under the IndiaAI Mission to build India’s sovereign large language model. That kind of work needs long-term investment.

Why sovereign AI is important

Sovereign AI means a country builds and controls important AI systems using local infrastructure, local talent and local priorities.

This matters for sensitive sectors like government, defence, healthcare, banking, education and public services. These sectors may not want all their data and AI workflows fully dependent on foreign systems.

For India, sovereign AI also means building models that understand local languages and culture. A model trained mainly on English internet data may not fully understand Indian context, legal documents, rural speech patterns or regional expressions.

Sarvam’s India-first approach fits this need.

Future possibilities

Sarvam’s future could move in a few directions.

It may become a core AI platform for Indian enterprises that want voice bots, translation, call-centre automation, document processing and Indian-language assistants.

It may also work closely with government departments and public-sector organizations where data security and language access are important.

Another possibility is device-level AI. If Sarvam’s models become efficient enough, they may run on phones, smart glasses, kiosks, customer service devices or edge systems.

For example, a small business owner could speak in Hindi and generate invoices. A health worker could record patient notes in a local language. A bank could offer voice support across Indian languages without forcing users into English.

That is the real promise of India-first AI.

Competitors and market landscape

Sarvam is competing in a difficult market.

Globally, it faces giants such as OpenAI, Google, Anthropic, Meta, Microsoft and Mistral. These companies have huge capital, talent and compute access.

In India, there are emerging AI players such as Krutrim, Soket AI Labs, Gnani.ai, CoRover, Reverie Language Technologies and several enterprise AI startups. Some focus on language models, some on voice, some on enterprise automation, and some on Indic language tools.

Sarvam’s edge is its full-stack approach. It is not only building one chatbot. It is building models, APIs, speech tools, translation systems and infrastructure for Indian use cases.

Still, the company will need to prove that its models are accurate, affordable and useful in real business environments.

Challenges ahead

The opportunity is large, but the challenges are also serious.

AI model development is expensive. Global competitors have deeper pockets. Enterprise customers need reliability, security and clear return on investment. Government use cases require trust, compliance and long sales cycles.

There is also the question of monetization. Many AI startups get attention for model releases, but the harder task is building sustainable revenue.

Sarvam will need to show that Indian businesses are willing to pay for its AI tools at scale.

Conclusion with key takeaways

Sarvam AI’s reported $75 million funding plan shows how important India-first AI has become. The company is trying to build models and tools that work for Indian languages, Indian users and Indian institutions.

Founded in 2023 by Vivek Raghavan and Pratyush Kumar, Sarvam has already raised major capital, released Indian-language AI models and been selected under the IndiaAI Mission for sovereign AI work.

If the new funding closes as reported, it can help Sarvam scale compute, product development, enterprise sales and future model training. The larger story is clear: India wants to become not only a user of AI, but also a builder of AI.

Key takeaways –

• Sarvam AI is reportedly set to raise $75 million from Nvidia, Glade Brook and others.
• The Bengaluru startup was founded in August 2023 by Vivek Raghavan and Pratyush Kumar.
• Sarvam builds AI models and tools for Indian languages, voice, translation, chat and document intelligence.
• The company earlier raised $41 million in Series A funding led by Lightspeed.
• Sarvam has been selected under the IndiaAI Mission to build India’s sovereign large language model.

Facts Input- ET