Baidu open-sources Unlimited OCR 2 for long documents

An open 3B OCR model that handles hundred-page documents in a single request.

Unlimited OCR 2 is Baidu's MIT-licensed 3B OCR model that parses hundred-page documents in one request; its SR-SWA attention keeps per-step decoding cost nearly flat and it scores 77.1 on MarathonDocBench.

Name

Unlimited OCR 2

Params

3B

License

MIT

Domain

software

Read it at MarkTechPost →

Coverage: 2 sources, in order

1. 27 Oct 05:26 MarkTechPost Read the article →
2. 27 Oct 08:30 Hugging Face Read the article →

← Front page