Yotta Data Services, a comprehensive digital transformation service provider, has partnered with AI startup Sarvam to build India’s first open-source foundational model, Sarvam 1. Sarvam 1 is the first LLM trained from scratch on an internal dataset of 4 trillion tokens by an Indian company, with compute in India, and provides efficient representation for 10 Indian languages. The Sarvam 1 model has been trained on Yotta’s Shakti Cloud infrastructure, delivering top-tier performance and reliability.
Utilizing Yotta’s Shakti Cloud, Sarvam has also launched products such as Sarvam Agents and developer-friendly APIs for businesses across various industries. The flagship product, Sarvam Agents, offers an advanced suite of tools that allows users to engage with AI-powered bots via telephone, WhatsApp, or in-app through natural, voice-driven conversations. Sarvam Agents are currently available in 10 Indian languages, including Hindi, Tamil, Telugu, Malayalam, Punjabi, Odia, Gujarati, Marathi, Kannada, and Bengali. These AI agents are designed to optimize various processes, including customer support, feedback collection, and employee engagement.
Powered by the NVIDIA accelerated computing platform with NVIDIA Hopper GPUs and the NVIDIA AI Enterprise software platform, Yotta’s Shakti Cloud helps companies train AI models faster and more efficiently, while also delivering consistent and seamless performance.
“Shakti Cloud provided Sarvam AI with flexible access to NVIDIA accelerated computing and NVIDIA AI Enterprise software to develop models with billions of data points to help them launch their services on time, in a seamless and efficient manner,” noted Sunil Gupta, Co-Founder, CEO and MD, Yotta Data Services. “This is just the beginning. We are excited about the potential this partnership holds and look forward to further boosting India’s AI capabilities.”
Through this partnership, Sarvam AI gets wider access to AI technology in India, helping ensure it maintains data sovereignty while also keeping data secure. Combined with a highly competitive pricing structure, Shakti Cloud will give Sarvam a significant competitive edge, while being backed by some of the most advanced AI infrastructure in the country.
"Partnering with Sarvam to create the Sarvam 1 foundational model, leveraging Yotta's Shakti Cloud infrastructure, marks a significant milestone in advancing India's AI landscape. Powered by cutting-edge chips and state-of-the-art infrastructure, Shakti Cloud accelerates AI model training, offering businesses the resources they need to drive innovation. This collaboration is focused on equipping Indian companies with world-class AI capabilities, enabling them to compete and excel in the global digital economy," said Darshan Hiranandani, Co-Founder and Chairman of Yotta.
"Our partnership with Yotta is a step towards bringing large-scale AI innovation in India," said Vivek Raghavan, Co-Founder, Sarvam. "Backed by Yotta’s advanced computing power, we can deliver reliable and affordable AI solutions to a wide range of industries. This collaboration allows us to offer a complete solution, leveraging NVIDIA AI Enterprise software, including NVIDIA NeMo for GPU-accelerated data processing pipelines and scalable and high-performance generative AI model training.”
Kari Briski, Vice President of AI Software at NVIDIA, said, “India is emerging as a global leader in AI innovation. Working with local technology experts like Yotta, NVIDIA supports the development of sovereign large language models and applications to help India’s businesses and people use AI to drive economic and social growth.”
Yotta is dedicated to making advanced AI technology available to more Indian businesses, with flexible pricing options that include pay-per-hour usage and long-term reservations. This move is part of Yotta’s mission to drive India’s digital transformation and support innovation across industries.
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.