# Is Unlimited OCR 2 SOTA?

Pending benchmarks. Announced Oct 27, 2026; no independent results yet.

As of Nov 3, 2026, Unlimited OCR 2 is pending benchmarks on isitsota: announced Oct 27, 2026, with no independent results on the boards we track yet.

Baidu's open-weight OCR vision-language model for long documents. Its SR-SWA attention condenses text-visual attention into a fixed-size set of working references, so it can parse all pages of a long document in one request while per-step decoding latency and memory stay nearly constant. Released under MIT on Hugging Face and GitHub, and served as a Baidu Cloud document-parsing API.

Canonical: https://isitsota.com/m/unlimited-ocr-2
Last updated: 2026-10-27
Tracking since: 2026-10-27

## Announced facts

| Fact | Value |
| --- | --- |
| Lab | Baidu |
| Announced | Oct 27, 2026 |
| API | Live · Baidu Cloud document parsing (Unlimited-OCR), an async task API based on Unlimited-OCR 2, free for a limited time |
| Open weights | Published · Hugging Face (baidu/Unlimited-OCR-2, Transformers with trust_remote_code) and ModelScope (PaddlePaddle/Unlimited-OCR-2); BF16 |
| Parameters | 3.34B (3,343,978,497 safetensors parameters) |
| Type | Vision-language OCR / document-parsing model (image-text-to-text) |
| Attention | Sparse Reference Sliding Window Attention (SR-SWA): text-visual attention compressed to a fixed-size set of working references (e.g. 512 tokens) |
| Multi-page parsing | infer_multi parses all pages of a document in one request with cross-page context, one global view per page; single pages use dynamic tiling (gundam mode) |
| License | MIT (model card and GitHub repo) |
| Serving | Transformers for quick tests; a customized SGLang wheel with an OpenAI-compatible API for serving |
| Benchmark released alongside | MarathonDocBench, a hundred-page document-parsing benchmark (baidu/MarathonDocBench on Hugging Face) |
| Predecessor | Unlimited-OCR (June 22, 2026), described in the arXiv paper 2606.23050; a separate model |

Tags marked unsure are not stated on the lab's own pages.

## What would let us call it

Verdicts use only independent boards we already track. Two consecutive snapshots are required before any crown.

- [x] Publicly announced with a source
- [x] Usable by the public (API or weights)
- [ ] Listed on an independent board we track
- [ ] Enough results for a rank in a category
- [ ] Holds across two snapshots (crowns only)

## What the lab claims

Self-reported. Shown for context; not used in any verdict.

- Lab claim: leading results on long-document parsing benchmarks, MPDocBench (dozens of pages) and MarathonDocBench (hundreds of pages).
- Lab claim: document parsing at virtually any length within a finite context window.

## Sources

- LAB: [Baidu on X: Unlimited OCR 2 is here, and it's open source](https://x.com/Baidu_Inc/status/2114910381311512987) (2026-10-27)
- REPO: [baidu/Unlimited-OCR-2 (Hugging Face model card)](https://huggingface.co/baidu/Unlimited-OCR-2) (2026-10-27)
- REPO: [baidu/Unlimited-OCR (GitHub)](https://github.com/baidu/Unlimited-OCR)
- LAB: [Document parsing (Unlimited-OCR) API, Baidu AI Cloud docs](https://cloud.baidu.com/doc/OCR/s/fmr1p39gb) (2026-10-27)

Tracking continues until an independent board lists it.

## Related

- Lab: [Baidu](https://isitsota.com/labs/baidu.md)
- All pending releases: https://isitsota.com/releases.md
- JSON: https://api.isitsota.com/v1/models/unlimited-ocr-2