Baidu Open Sources Unlimited OCR 2 Document Parser

Oct 28, 2026
Unlimited OCR 2 uses sparse attention to parse documents with thousands of pages while keeping decoding memory and latency nearly constant.

Baidu released Unlimited OCR 2, an open source model designed to parse documents of any page count in one pass, according to Pandaily. Its code and weights are available on GitHub and Hugging Face under the MIT license, while hosted access is available through Baidu AI Cloud and the Baidu AI Open Platform.

The model uses Sparse Reference Sliding Window Attention, which selects relevant pages and visual tokens instead of applying full attention across an entire document. Baidu says this keeps decoding latency and memory nearly constant as document length increases. In one demonstration, the model processed more than 3,600 pages across all seven Harry Potter books in a single run.

Baidu reports scores of 84.6 on MPDocBench and 77.1 on its new MarathonDocBench benchmark for documents containing hundreds of pages. In tests with 100 page documents, the company says the new attention method decoded each step seven times faster than full attention and 5.6 times faster than the previous design. Independent evaluations were not available at publication.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Daily AI Brief.

Also, consider following us on social media:

AI Brief AI Brief (X)
Free newsletter

Daily AI Brief

Daily report covering major AI developments and industry news, with both top stories and complete market updates

Guide
Private humanoid makers by latest disclosed equity round

Private humanoid makers by latest disclosed equity round

Thirteen independent humanoid robot makers, ordered by the size of their latest disclosed round, checked on 21 October 2026

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.