India’s First Vision-Language Document Foundation Model

Patram - India’s First Vision-Language Document Foundation Model

Indic knowledge isn’t on the web—it’s in documents. From government forms and academic papers to legal records and regional literature, India’s richest, most refined textual data lives in books and scanned files. Yet today, open-source document understanding for Indic languages is nearly nonexistent.

Imagine the impact:

  • A farmer in Bihar can ask questions about a government scheme—right from a scanned notice.
  • A student in Kerala can get summaries of a textbook written in Malayalam.
  • Lawyers can search judgments in regional languages, with citations.
  • Banks can verify KYC documents automatically, regardless of language or format.

Key Capabilities of Patram

Patram isn’t just smart—it’s made for Bharat.

Multilingual & Multimodal

Multilingual & Multimodal

Understands documents in Indian languages and English, from scanned images to complex layouts.

Smart Understanding

Smart Understanding

Responds to natural language prompts—summarize, extract, answer questions, and more.

Made for India, in India

Made for India, in India

Trained on real Indic documents for high relevance across law, governance, education, finance, etc.

Why Choose Patram?

BharatGen proudly launches Patram: India’s first Vision-Language Document Foundation Model—unlocking Indic documents and powering the next era of Indic Document Intelligence.

You can now access the model at the following links:

Explore Patram

Power your document understanding apps with Patram—fast, accurate, and multilingual intelligence for complex Indian documents.

Partner with Us

Join BharatGen to unlock insights and transform document workflows across India with next-gen AI solutions.

Scroll to Top