Models/OpenHathi
OpenHathi mark

OpenHathi

Fine-tunedOpen-sourceMultilingualहिन्दीEnglish

In plain words

OpenHathi is a 7B-parameter Hindi-first model built on the LLaMA architecture by Sarvam AI and AI4Bharat. It was one of the first publicly available Hindi LLMs.

How to use it

🚧
This section is coming soon.
A step-by-step guide for OpenHathi will appear here soon.

Languages & scripts supported

Works in these languages, in both native script and Roman typing.

हिन्दी HindiEnglish English+ Roman & code-mixed

Strengths & limits

An honest look at what it does well and where it struggles.

Good at
Native language processing
Text generation
Where it struggles
!Highly complex code generation
!Low-resource dialects without fine-tuning

About the maker

Who builds and maintains OpenHathi.

Sarvam AI / AI4Bharat logo
Sarvam AI / AI4Bharat
1 models

This entry credits a collaboration between Sarvam AI, a startup founded by former AI4Bharat researchers that has since raised significant funding and been selected to help build India's sovereign LLM, and AI4Bharat, the IIT Madras lab it grew out of. Together they released OpenHathi-Hi, one of the first openly available Hindi large language models.