All tasks / Models
Unlimited-OCR
Baidu's 3B open-weights OCR model built for one-shot parsing of long multi-page documents.
- Baidu
- Open weights
- Released 22 Jun 2026
Where it ranks
- SDocuments & OCRModels · Local · 94 on OmniDocBench v1.6 (end-to-end, v1.6_full)
Facts
- Vendor
- Baidu
- Family
- Unlimited-OCR
- Released
- 22 Jun 2026
- Weights
- Open: download and run it yourself
- License
- MIT
- Parameters
- 3.34B
- GPU memory to run locally
- ~8 GB (Estimate, no official figure: BF16 weights ≈ 3.34 B × 2 GB + ~1.5 GB runtime overhead. Size is the Hugging Face safetensors total.)
- Access
- Free (open weights)
- Last verified
- 19 Oct 2026
Price from github.com, 19 Oct 2026.
Benchmark results
| Benchmark | Result | Entry on the leaderboard | As of | Source |
|---|---|---|---|---|
| MDPBench (multilingual document parsing) | 54.1% | Unlimited-OCROverall (public split, page-level mean of text, formula and table scores) from the board's leaderboard.json (updated 2026-10-04). | 4 Oct 2026 | huggingface.co |
| OmniDocBench v1.6 (end-to-end, v1.6_full) | 94 | Unlimited-OCROverall score in the README's end-to-end table 'OmniDocBench (v1.6_full)' (last leaderboard update 2026-09-11). | 11 Sep 2026 | github.com |
| ParseBench (models) | 46.2% | Unlimited-OCROverall (mean of tables, charts, content faithfulness, semantic formatting, visual grounding) from leaderboard.csv in the ParseBench repo (last leaderboard commit 2026-10-09). | 9 Oct 2026 | github.com |
Where you can use it
No tracked tool lists this model yet; as open weights it runs in local runtimes.