{
 "schema_version": 1,
 "generated": "2026-10-30",
 "repo": "baidu/Unlimited-OCR",
 "name": "Unlimited-OCR",
 "owner": "baidu",
 "slug": "baidu-unlimited-ocr",
 "github": "https://github.com/baidu/Unlimited-OCR",
 "what": "Unlimited OCR 2 is Baidu's open OCR model for ultra-long document parsing. Sparse Reference Sliding Window Attention lets it parse documents of any length within a finite context window, with multi-page inference that uses cross-page context. It runs with Hugging Face Transformers for quick tests and a customized SGLang server for serving, and has a Hugging Face Spaces demo.",
 "why": "30k stars, talked about in 3 AI sources we track",
 "language": "Python",
 "requirements": [
  "An NVIDIA GPU (tested on an A800 80GB) with CUDA",
  "Python 3.10 with torch 2.13.0 and transformers 4.52.1, or uv for the SGLang setup",
  "PyMuPDF to convert PDFs to page images"
 ],
 "audience_note": "ML engineers and data teams who need to turn long scanned PDFs and reports into text at scale.",
 "risk": "none",
 "section": "earlier",
 "weekly_rank": null,
 "category": "Data and analytics",
 "url": "https://dedupedai.com/use-cases/baidu-unlimited-ocr/",
 "use_case_count": 3,
 "use_cases": [
  {
   "id": "baidu-unlimited-ocr--parse-a-long-multi-page-document-with-unlimited-ocr-2",
   "title": "Parse a long multi-page scan with cross-page context",
   "best": true,
   "url": "https://dedupedai.com/use-cases/baidu-unlimited-ocr/parse-a-long-multi-page-document-with-unlimited-ocr-2/",
   "json": "https://dedupedai.com/use-cases/u/baidu-unlimited-ocr/parse-a-long-multi-page-document-with-unlimited-ocr-2.json"
  },
  {
   "id": "baidu-unlimited-ocr--serve-unlimited-ocr-2-with-sglang-for-batch-pdf-parsing",
   "title": "Serve Unlimited OCR 2 with SGLang for hundred-page PDFs",
   "best": false,
   "url": "https://dedupedai.com/use-cases/baidu-unlimited-ocr/serve-unlimited-ocr-2-with-sglang-for-batch-pdf-parsing/",
   "json": "https://dedupedai.com/use-cases/u/baidu-unlimited-ocr/serve-unlimited-ocr-2-with-sglang-for-batch-pdf-parsing.json"
  },
  {
   "id": "baidu-unlimited-ocr--benchmark-ocr-on-marathondocbench",
   "title": "Compare your OCR pipeline on the MarathonDocBench ultra-long benchmark",
   "best": false,
   "url": "https://dedupedai.com/use-cases/baidu-unlimited-ocr/benchmark-ocr-on-marathondocbench/",
   "json": "https://dedupedai.com/use-cases/u/baidu-unlimited-ocr/benchmark-ocr-on-marathondocbench.json"
  }
 ],
 "disclaimer": "Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk."
}