---
format: "aidr-story-markdown/v1"
id: "ff02ebddda4a7c855c44ca9b49917eaaa54176c4de5c821d09ec10117dd1f781"
canonical_url: "https://aidr.today/ff02ebdd?lang=en"
title: "Baidu Open-Sources Unlimited OCR 2 for Hundred-Page Document Parsing"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-27T05:26:36.000Z"
category: "Open Source"
topics: ["open-source"]
source_urls: ["https://www.marktechpost.com/2026/10/26/baidu-unlimited-ocr-2-open-source-long-document-ocr/","https://huggingface.co/baidu/Unlimited-OCR-2"]
summary: "Baidu has released Unlimited OCR 2, an MIT-licensed 3B OCR model that parses hundred-page documents in one request. Its SR-SWA attention keeps per-step decoding cost nearly flat, scoring 77.1 on MarathonDocBench."
---

# Baidu Open\-Sources Unlimited OCR 2 for Hundred\-Page Document Parsing

> [Open the canonical story](<https://aidr.today/ff02ebdd?lang=en>)

**Published:** 2026-10-27T05:26:36.000Z
**Category:** Open Source
**Topics:** open\-source

## Summary

Baidu has released Unlimited OCR 2, an MIT\-licensed 3B OCR model that parses hundred\-page documents in one request\. Its SR\-SWA attention keeps per\-step decoding cost nearly flat, scoring 77\.1 on MarathonDocBench\.

## Sources

- [Story source](<https://www.marktechpost.com/2026/10/26/baidu-unlimited-ocr-2-open-source-long-document-ocr/>)
- [Story source](<https://huggingface.co/baidu/Unlimited-OCR-2>)