{"id":"unlimited-ocr-2","slug":"unlimited-ocr-2","name":"Unlimited OCR 2","orgId":"baidu","family":"unlimited-ocr-2","releaseDate":null,"availability":"announced","openWeights":true,"license":null,"paramsTotal":null,"paramsSource":null,"contextWindow":null,"knowledgeCutoff":null,"modelCardUrl":"https://x.com/Baidu_Inc/status/2114910381311512987","hfRepo":null,"modalities":["image","text"],"aliases":["Unlimited OCR 2","unlimited ocr 2","unlimited-ocr-2","unlimited-ocr 2","baidu/Unlimited-OCR-2","unlimited ocr works 2","wenxin unlimited-ocr 2"],"aliasNote":null,"origin":"tracked","tracked":{"trackedSince":"2026-10-27","announced":"2026-10-27","blurb":"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.","api":"live","openWeights":"yes","params":{"text":"3.34B (3,343,978,497 safetensors parameters)","source":"Hugging Face model metadata (baidu/Unlimited-OCR-2)"},"facts":[{"label":"Type","value":"Vision-language OCR / document-parsing model (image-text-to-text)"},{"label":"Attention","value":"Sparse Reference Sliding Window Attention (SR-SWA): text-visual attention compressed to a fixed-size set of working references (e.g. 512 tokens)"},{"label":"Multi-page parsing","value":"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)"},{"label":"Weights","value":"Hugging Face (baidu/Unlimited-OCR-2, Transformers with trust_remote_code) and ModelScope (PaddlePaddle/Unlimited-OCR-2); BF16"},{"label":"License","value":"MIT (model card and GitHub repo)"},{"label":"Serving","value":"Transformers for quick tests; a customized SGLang wheel with an OpenAI-compatible API for serving"},{"label":"API","value":"Baidu Cloud document parsing (Unlimited-OCR), an async task API based on Unlimited-OCR 2, free for a limited time"},{"label":"Benchmark released alongside","value":"MarathonDocBench, a hundred-page document-parsing benchmark (baidu/MarathonDocBench on Hugging Face)"},{"label":"Predecessor","value":"Unlimited-OCR (June 22, 2026), described in the arXiv paper 2606.23050; a separate model"}],"claims":["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":[{"kind":"lab","title":"Baidu on X: Unlimited OCR 2 is here, and it's open source","url":"https://x.com/Baidu_Inc/status/2114910381311512987","date":"2026-10-27"},{"kind":"repo","title":"baidu/Unlimited-OCR-2 (Hugging Face model card)","url":"https://huggingface.co/baidu/Unlimited-OCR-2","date":"2026-10-27"},{"kind":"repo","title":"baidu/Unlimited-OCR (GitHub)","url":"https://github.com/baidu/Unlimited-OCR"},{"kind":"lab","title":"Document parsing (Unlimited-OCR) API, Baidu AI Cloud docs","url":"https://cloud.baidu.com/doc/OCR/s/fmr1p39gb","date":"2026-10-27"}],"firstBenchmark":null},"organization":{"id":"baidu","name":"Baidu","slug":"baidu","country":null,"url":"https://www.baidu.com","color":"#A39E8F","mark":"BA"},"price":null,"answer":{"word":"Pending benchmarks.","scope":"Announced Oct 27, 2026; no independent results yet.","lead":"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.","confidence":"low"},"status":"PENDING_BENCHMARKS","verdicts":[]}