Avataar AI launches Varya, a culturally aware video model built for India's scale
Avataar AI, one of 12 startups selected under India's AI Mission, introduces Varya – a video generation model based on Alibaba's open-source Wan 2.2. Using distillation, Varya runs 10x faster and costs 20x less than competitors, with prices as low as $0.005 per second of video.
India's AI development has lagged behind the US, Europe, and China, with only a handful of startups releasing models, mostly large language or voice models. To spur progress, the government launched the India AI Mission, a roughly $1.2 billion initiative that provides selected startups with subsidized GPU compute in exchange for releasing their models publicly.
One of the 12 selected startups, Peak XV-backed Avataar AI, has launched Varya, a video model designed to understand local context—recognizing different festivals, food, and clothing. The company, which builds video tools for e-commerce, did not create Varya from scratch. It started with Wan 2.2, a publicly available video generation model from Alibaba, and used a technique called distillation to compress its capabilities into a leaner, faster version optimized for Avataar's use cases.
The result: Varya runs in four steps instead of Wan 2.2's 50, producing video 10 times faster and at a fraction of the cost. On an Nvidia H200 GPU, Varya generates a five-second 720p clip in 45 seconds, compared to 1,230 seconds for Wan 2.2. The most striking aspect is the price: the company plans to charge ₹0.48 ($0.005) per second of video on its hosted service, roughly 20 times cheaper than models like Veo, Kling, Luma, and Runway, which typically cost $0.10 or more per second.
“India is a video-first market. We see this across every large consumer internet product in India: video wins over text. Current AI video models are too expensive for population-scale use in India. If video AI is going to reach students, teachers, MSMEs, creators, enterprises, and public services, costs have to come down dramatically. Cost is the biggest unlock for AI adoption in India,” said Rajan Anandan, managing director at Peak XV.
Image and video generation models often miss cultural nuances and produce stereotyped outputs. Avataar AI says it used curated data to train Varya to recognize cultural nuances including food, clothing, architecture, and festivals. Varya will be released as an open-weight model on India's AIKosh portal—the government's centralized repository for AI models and datasets—along with its training data, meaning developers can self-host or modify it. Avataar also plans to make the model available to its enterprise customers and is open to partnerships with video tools like Higgsfield and Adobe Firefly. Anyone can try it now on its website using text prompts or reference images.
Varya's launch reflects a fundamental tradeoff in India's AI ambitions. Industry veterans have noted that India can make its mark by creating applications and a robust developer ecosystem rather than competing on foundation models, given the lack of compute and limited quality data. The India AI Mission is part of a broader government push to close that gap. Last year, it selected 12 startups—including Avataar AI—to develop AI models with cost-efficient compute. Earlier this year, IT minister Ashwini Vaishnaw said India aims to attract $200 billion in AI investment by 2028 and more than double its GPU capacity within six months.
