Indra OCR 1.1 is out now! - 1M param main OCR model, 44 pages/s on...
Vik Paruchuri@VikParuchuri
34 views
Oct 21, 2026
~1 min read
Advertisement
1
Indra OCR 1.1 is out now!
- 1M param main OCR model, 44 pages/s on CPU
- Within 4 points of 3.4B param DeepSeek OCR 2 on OmniParseBench
- 15x faster than Tesseract, and more accurate
- 80+ languages, plus handwriting
- 1M param main OCR model, 44 pages/s on CPU
- Within 4 points of 3.4B param DeepSeek OCR 2 on OmniParseBench
- 15x faster than Tesseract, and more accurate
- 80+ languages, plus handwriting
0:29
2
Get it here - huggingface.co/datalab-to/ind…
Code at github.com/datalab-to/ind…
Read more at datalab.to/blog/indra-ocr…
Code at github.com/datalab-to/ind…
Read more at datalab.to/blog/indra-ocr…
3
Indra is close to several much larger models on OmniParseBench, including Deepseek OCR 2, when used with the --math flag. It beats the 1.1B param HunyuanOCR (66.7 vs 64.6).
The --math flag uses a larger parameter model just for math (12M total params).
The --math flag uses a larger parameter model just for math (12M total params).
4
The main changes from 1.0 are improved language support, handwriting (this is limited and mostly English for now), and fixes for different rotation and layout issues.
5
I hope it works well for you. We're continuing to improve Indra OCR - let us know if you have any feedback.