Context
Recently, Avataar.ai launched Varya AI, a video-generation model developed to deliver cost-effective AI-powered video creation and support wider adoption of generative AI applications in India.
About Varya AI
- Varya AI is a distilled video-generation model designed to make AI-powered video creation more affordable and accessible in India.
- It has been developed with support from the India AI Mission.
- The model uses distillation, a machine-learning technique in which a compact model learns from a larger one, reducing video generation from nearly 50 processing steps to just 4 while maintaining comparable output quality.
- It can generate videos at about ₹0.48 per second, making it up to 10 times more cost-efficient than several leading global models.
- Users can create videos through text prompts or image inputs and extend them into longer video sequences.
- Its applications span education (visual learning content), MSMEs (product advertising), and governance (video-based dissemination of public information).
- By significantly lowering computational requirements, the model enables faster, scalable, and cost-effective video generation.
