Model reference · open weights

indic-ocr

LLMs bodhan-ai Image→text 1 build Its own licence terms 626 dl/mo

indic-ocr is an open-weight language model from bodhan-ai. indic-ocr (BF16) weighs 1.9 GB; the smallest configuration that runs it is RTX 3060 12 GB.

IndicOCR is an image-to-text model developed by bodhan-ai for parsing document pages into reading-ordered Markdown, LaTeX math, and HTML or Markdown tables.
It supports printed text in English and 22 Indian languages, as well as handwriting in English and 12 Indian languages.
The system consists of a 33 million parameter layout detector and a 0.8 billion parameter OCR module, and is released under the Indic Open Model License v1.0.

Summary of the bodhan-ai/indic-ocr model card, 2026-10-01

What it is

What it runs on

Memory and cards for indic-ocr (BF16)

1.9 GBweights, file size

762 MBruntime overhead, at least

How much memory each request adds isn't estimated yet for this architecture. The weights need at least the cards below, plus room for the context.

How it works

Details

Languages

enasbnbrxdoiguhiknkskokmaimlmnimr