{"apiVersion":"v1","taxonomyVersion":1,"generatedAt":"2026-10-29T15:48:20.452Z","data":{"recordId":"wsa_e4xixqvv53t5","score":1,"confidence":"high","capped":false,"capReason":null,"sources":[{"n":1,"url":"https://huggingface.co/baidu/Unlimited-OCR-2","publisher":"Hugging Face","authority":"owner-controlled","verification":"verified","lastRetrievedAt":"2026-10-28T00:19:28.484Z","detectedPublishedAt":"2026-10-26T03:57:46.000Z","rights":"original"}],"claims":[{"id":"9314afec-4461-403b-9d6f-e2a9a9e7b2dd","kind":"capability","statement":"SR-SWA compresses text-visual attention interactions to a fixed-size set of working references, such as 512 tokens.","citations":[{"source":1,"supportProbability":0.98,"supports":true,"quote":{"text":"* Sparse Reference Sliding Window Attention (SR-SWA): Compresses text-visual attention interactions from ultra-long sequences down to a fixed-size set of working references (e.g., 512 tokens).","rights":"original"}}]},{"id":"96be79f7-7a4f-4293-84ca-a7a845001a48","kind":"capability","statement":"The model supports multi-page inference, with all pages of a document parsed in one request.","citations":[{"source":1,"supportProbability":0.98,"supports":true,"quote":{"text":"All pages of a document are parsed in one request, so the model can use cross-page context.","rights":"original"}}]},{"id":"39742621-0ada-4fcd-b9fe-48aa04d99e40","kind":"capability","statement":"The release describes unlimited-horizon parsing within a finite context window.","citations":[{"source":1,"supportProbability":0.97,"supports":true,"quote":{"text":"* From Long- to Unlimited-Horizon Parsing: Powered by SR-SWA, Unlimited-OCR 2 achieves true, unbounded document parsing within a finite context window—effectively eliminating quadratic attention scaling and memory bottlenecks as documents grow.","rights":"original"}}]},{"id":"db51ccf0-918a-4ec7-b7b7-cf9c7471efef","kind":"availability","statement":"The model is available on Hugging Face, ModelScope, and Baidu Cloud.","citations":[{"source":1,"supportProbability":0.76,"supports":true,"quote":{"text":"It is now available at [Hugging Face Spaces](https://huggingface.co/spaces/baidu/Unlimited-OCR).","rights":"original"}}]},{"id":"cd613c83-caa0-428c-b386-ee1ec7253481","kind":"deployment","statement":"The model supports Transformers and SGLang inference.","citations":[{"source":1,"supportProbability":0.98,"supports":true,"quote":{"text":"- [2026/06/28] 🤝 Thanks to the [vLLM community](https://github.com/vllm-project/vllm) and [Tianyu Guo](https://github.com/gty111) for their support, our model now supports vLLM inference.","rights":"original"}}]},{"id":"425c8bd4-9ebe-48c6-b0e1-19ecec4b9eaf","kind":"recommendation-relevant","statement":"The model is released under the MIT licence.","citations":[{"source":1,"supportProbability":0.75,"supports":true,"quote":null}]}]},"nextCursor":null,"license":{"catalog":{"name":"CC BY 4.0","spdx":"CC-BY-4.0","url":"https://creativecommons.org/licenses/by/4.0/","attribution":"wowsuchai","attributionUrl":"https://wowsuchai.withglint.app","covers":["catalog metadata","classifications","generated summaries","feed event metadata"]},"excluded":{"covers":["source quotations","linked source content"],"terms":"Not licensed by wowsuchai. Source quotations and linked source content keep their original rights."}}}