Avataar Launches Varya to Make AI Video Creation Affordable in India

Developed with support from the IndiaAI Mission and subsidized national AI compute infrastructure, Varya represents India’s push toward indigenous foundation models that can operate at population scale.

Avataar, an AI-native transformation company, has unveiled Varya, a compact video AI model designed specifically for India, to make cutting-edge video creation affordable and accessible to a broad range of users. The launch event in New Delhi was attended by Shri S. Krishnan, Secretary of the Ministry of Electronics and Information Technology, highlighting the government's support for homegrown AI innovations. Varya has been built to understand the cultural richness of India's regions, festivals, communities, food, clothing, public spaces, and daily life, allowing anyone—from teachers in village classrooms to small businesses creating product advertisements—to turn simple ideas into fully realized video stories.

---

Distilled Technology Makes AI Efficient and Cost-Effective

Varya leverages distilled video generation, a machine learning technique that compresses a large, slow "teacher" model into a smaller, faster "student" model that replicates the same output quality with far less computation. While traditional video models require over fifty iterative steps to generate a single clip, Varya can achieve comparable results in just four steps, making video production dramatically faster and more affordable. Internal benchmarks show the model can generate video at just ₹0.48 per second, making it up to ten times more cost-efficient than several leading global video models, which ensures AI video creation is accessible to the wider population, not just a privileged few.

---

Indigenous AI for India's Next Billion Stories

Developed with support from the IndiaAI Mission and subsidized national AI compute infrastructure, Varya represents India's push toward indigenous foundation models that can operate at population scale. CEO Sravanth Aluru emphasized that affordability is critical for inclusion in a country of 1.4 billion people, and Varya demonstrates that high-quality video AI can be both efficient and context-aware, addressing India's diverse needs. By allowing users to transform ideas into video stories with just a few prompts, the model opens doors for lessons, advertisements, public information, and personal content that reflect the nation's culture and everyday life, demonstrating that efficient, accessible AI can fuel creativity at scale.