via TechCrunch AI
Cheaper, Faster, and Culturally Aware: Avataar’s Video AI Targets India’s Scale
India’s AI model output has lagged behind the U.S., Europe, and China, with only a handful of startups releasing models, primarily large language or voice models. To spur development, the Indian government launched the India AI Mission, a roughly $1.2 billion initiative that, among other things, provides selected startups with subsidized GPU compute in exchange for public model releases. One of the 12 chosen startups, Avataar AI, has introduced Varya, a new video model purpose-built to understand local cultural context—such as identifying festivals, food, and clothing.
Backed by Peak XV, Avataar focuses on creating video tools for e-commerce. Varya was not built from scratch; it was distilled from Alibaba’s publicly available Wan 2.2 video generation model. Distillation compresses the model’s capabilities into a leaner, faster version optimized for Avataar’s specific use cases. The result? Varya runs in just four steps compared to Wan 2.2’s 50, generating video 10 times faster at a fraction of the cost.
Concretely, on an NVIDIA H200 GPU, Varya can produce a 5-second 720p clip in 45 seconds, versus 1,230 seconds for Wan 2.2. Pricing is equally striking: Avataar plans to charge ₹0.48 ($0.005) per second of video on its hosted service—roughly 20 times cheaper than competitors like Veo, Kling, Luma, and Runway, which typically charge $0.10 or more per second.
“India is a video-first market. We see this across every large consumer internet product: video wins over text. Current AI video models are too expensive for population-scale use. If video AI is to reach students, teachers, MSMEs, creators, enterprises, and public services, costs must fall 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, producing stereotyped or generic outputs—a problem TechCrunch has previously reported. Avataar says it used curated data to train Varya to recognize cultural elements including food, clothing, architecture, and festivals.
Varya will be released as an open-weight model on India’s AI Kosh portal—the government’s centralized repository for publicly available AI models and datasets—along with its training data. This allows developers to self-host or modify it for their own needs. Avataar also plans to make the model available to enterprise customers and is open to partnerships with video tools like Higgsfield and Adobe Firefly. Anyone can try it now on Avataar’s website using text prompts or reference images.
Varya’s launch reflects a fundamental tradeoff in India’s AI ambitions. Industry veterans note that India can make its mark by creating applications and a robust developer ecosystem rather than competing on foundational model scale alone.