Feature Request: Supports Unlimited-OCR

[Leroy-X]

Prerequisites

• I am running the latest code. Mention the version if possible as well.
• I carefully followed the README.md.
• I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
• I reviewed the Discussions, and have a new and useful enhancement to share.

Feature Description

https://github.com/baidu/Unlimited-OCR
https://huggingface.co/baidu/Unlimited-OCR

Motivation

The longer the document becomes, the more memory the model consumes and the slower it gets.

Baidu’s new Unlimited OCR project tackles this challenge with an interesting idea: instead of remembering everything it has generated, the model learns to remember only what is necessary.
The result is an OCR system capable of parsing dozens of pages in a single pass while maintaining nearly constant memory usage.

https://medium.com/data-science-in-your-pocket/baidus-unlimited-ocr-beats-deepseek-ocr-parses-entire-book-in-one-go-6e3e1a8c9b34

Possible Implementation

No response

[o7si]

Related PRs already in progress: #24969 (merged) and #24975.