India does not intend to lag behind in developing basic AI models and the models developed under IndiaAI’s mission are an answer to those who question India’s potential to develop advanced AI platforms, a senior government official said on Friday.
While startup Avataar.Ai launched the country’s first video infrastructure model, Electronics and IT Secretary S Krishnan said the government is supporting a wide range of infrastructure models across various applications, not just language models.
“For those who say modeling and these kinds of initiatives can’t succeed in India, this (Avataar.Ai’s video platform) is the answer to show that it is indeed possible and the answer to show that India is second to none in this kind of endeavor,” Krishnan said.
Avataar.Ai has launched Varya, a unique AI video model developed with support from IndiaAI Mission, to make high-quality video generation more affordable, accessible and relevant for India.
Users can upload photos or raw videos to the platform and generate edited versions by simply describing the desired video output details.
Avatar CEO and co-founder Sravanth Arul said users will be able to create 211 seconds of videos for every Rs 100 spent on the platform.
He claimed that Varya reduces video generation time from 50 steps to 4 steps, making it 10 times more efficient than leading models.
“Today, we are making video generation approximately 27 times faster and 27 times cheaper as well. We achieved this because we collectively believe that affordable AI is critical if we truly want to unlock India’s productivity curve,” said Aluru. Aluru said the company has developed a video AI platform with 14 billion parameters.
“We anticipate that we’re going to open up a market that was unserved yesterday. We’re going to probably create a new market for people who weren’t doing video because they thought it was too expensive and a hassle to actually start doing it. We need to look at the overall impact and evolve,” he said. When asked about the impact on work in the video development sector, Aluru said people’s skills need to improve and AI tools need to work with human ingenuity.
“We’re allowing the ideas and imagination of creators, and therein lies the value. We can transform them into actual, physical videos that can be viewed in an efficient way. We remove the grunt work, but preserve the ability for someone to develop great ideas that they haven’t thought of before. I think there’s still plenty of originality and human ingenuity out there,” Arles said.
