sarvam ai

India must urgently develop its own frontier-scale AI models to avoid becoming a passive consumer and to shape global technology rules, according to sarvam ai co-founder Pratyush Kumar. The company is preparing to train a trillion-parameter AI model, emphasizing the need for strategic capability and ownership of foundational models to capture economic value and prevent repeating past technological revolution mistakes.

Former NITI Aayog member Arvind Virmani advocates for India to leverage Artificial Intelligence to boost human productivity, not replace workers. He emphasizes combining AI with India's vast workforce, focusing on practical applications in healthcare, education, and agriculture, rather than solely pursuing robotics. Virmani also stresses early AI education integration and government support for AI research and development.

Senior Vice President Dino Bekis explains the architecture behind a partnership that doubles as the opening move in Qualcomm's broader India ambition

sarvam ai has launched two new AI models, Sarvam 30B and Sarvam 105B, as open-source. These models were developed from scratch in India, trained on local data, and utilize IndiaAI Mission computing power. They are designed for speed and complex tasks respectively, supporting all 22 Indian languages and voice interactions. These models show strong performance on various benchmarks.

A lean team of 15 researchers, many in their twenties, at Sarvam successfully built a 105-billion-parameter foundational LLM from scratch. Spearheaded by Rahul Aralikatte, the young team managed data engineering, pre-training, and safety guardrails, showcasing India's emerging AI talent. Sarvam, a full-stack GenAI company, aims to build technology for India's specific needs.

Google CEO Sundar Pichai lauded sarvam ai for its advancements in local AI models tailored for Indian languages and contexts. The startup's AI model reportedly outperforms major players like Google's Gemini and OpenAI's ChatGPT in specific benchmarks, showcasing its potential for India's digital future.

Bengaluru-based sarvam ai has unveiled Sarvam Kaze, its AI-powered smart glasses designed and built in India. Showcased at the India AI Impact Summit 2026, these glasses aim to cater to Indian languages and use cases, differentiating themselves from English-centric competitors.

Sarvam, an Indian AI startup, has launched its Indus chat app. This app is designed for Indian languages and users. It enters a competitive generative AI market. Google CEO Sundar Pichai praised Sarvam's work on local AI models. The Indus app is available on web and mobile platforms. It uses Sarvam's 105B large language model.

Bengaluru-based sarvam ai is making significant strides in India's AI landscape, developing foundational models tailored for Indian languages. Their OCR tool, Sarvam Vision, leads benchmarks, while Bulbul V3 offers natural-sounding text-to-speech across multiple Indian languages, proving competitive and affordable against global players.

sarvam ai has launched its Sarvam Edge models, offering on-device AI for Indian languages that compete with global players. These models provide offline speech recognition, synthesis, and translation, running entirely on consumer devices. This approach ensures user data privacy, eliminates per-query costs, and makes AI accessible even in areas with poor internet connectivity.

Sridhar Vembu highlights sarvam ai's achievement in developing world-class AI affordably and sustainably with lower energy and compute needs. He believes software code generation can adopt more efficient AI techniques, crucial as the planet cannot sustain current AI energy demands. This approach balances performance with environmental responsibility amid growing concerns.

Bengaluru-based sarvam ai raises $41 million in series A funding led by Lightspeed, with participation from Peak XV Partners and Khosla Ventures. The funding highlights the strong investor interest in AI startups in India. sarvam ai focuses on generative AI and is developing a full stack for training custom AI models and deploying them. The company plans to collaborate with Indian enterprises to build domain-specific AI models and aims to create population-scale impact for public-good applications. The startup will prioritize catering to India's unique needs, including supporting local languages and voice-first interfaces.

sarvam ai is likely to be the first recipient of funding from the IndiaAI Mission, potentially receiving Rs 220 crore in the form of access to Nvidia H100 chips. This support aims to bolster Sarvam's efforts in developing an indigenous large language model, particularly one tailored for Indian languages, aligning with the mission's goals.

Indian AI startup Sarvam has launched two powerful large language models, built from the ground up for Indian languages. These models, boasting 30 and 105 billion parameters respectively, are designed to support all 22 official Indian languages and are optimized for voice interaction. This move directly challenges global AI giants in India's rapidly expanding AI market.

Indian startup Sarvam has launched a 105-billion-parameter large language model, performing on par with global counterparts and outperforming others on Indian language benchmarks. This homegrown AI model, developed with government support, signifies India's capability to build state-of-the-art AI from scratch, targeting commercial applications.