Skip to content
← Models
Open weightsSarvam AI/10 trustVerified

Sarvam-1

India's open-source small LLM for Indic languages — 10 Indian languages plus English, 4-6x faster inference.

Visit official page →

About

Sarvam-1, launched in late 2024 by Bengaluru-based Sarvam AI, is the first open-source LLM purpose-built for Indian languages. It's a small language model (under 10 billion parameters) supporting ten Indian languages — Bengali, Gujarati, Hindi, Kannada, Malayalam, Marathi, Oriya, Punjabi, Tamil, Telugu — plus English, optimized for production deployment with 4-6x faster inference than conventional LLMs of comparable capability.

Sarvam AI was selected by the Government of India under the IndiaAI Mission (April 2025) to build the country's sovereign 120-billion-parameter open-source foundation model for governance and public services — a project distinct from Sarvam-1 itself, which remains the practical open model available today. The company has raised around $50 million from Khosla Ventures, Lightspeed, and Peak XV. Note: exact launch day within October 2024 is approximate in public reporting.

Specs

Context window

Pricing

Free quota

API access

Weights

Open weights

Released

October 1, 2024

Benchmarks

Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.

Use cases

indic-nlpvoice-aiedge-deploymentgovernance

Trust score — 8/10

/10 trustVerified

Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 59; Trending score 76/100 from researcher assessment.

More in the sarvam family

Sources