{"id":"youtu-parsing-omni","slug":"youtu-parsing-omni","name":"Youtu-Parsing-Omni","orgId":"tencent","family":"youtu-parsing-omni","releaseDate":null,"availability":"ga","openWeights":true,"license":"Custom License Terms of Youtu-Parsing (HF tag other, license_name youtu-parsing)","paramsTotal":5000000000,"paramsSource":"slug","contextWindow":null,"knowledgeCutoff":null,"modelCardUrl":"https://huggingface.co/tencent/Youtu-Parsing-Omni","hfRepo":null,"modalities":["text","image","audio","video"],"aliases":["Youtu-Parsing-Omni","Youtu Parsing Omni","tencent/Youtu-Parsing-Omni","youtu-parsing-omni-5b"],"aliasNote":null,"origin":"tracked","tracked":{"trackedSince":"2026-11-08","announced":"2026-10-10","blurb":"Tencent Youtu Lab's roughly 5B omni-modal parser converts document pages, images, charts, flowcharts, geometry, audio and video into structured OmniSchema JSON. A shared Youtu-Omni-Encoder handles perception across modalities, with modality-specific stems/projectors and a text decoder. Weights and self-hosting examples are public; no public hosted API is confirmed.","api":"none","openWeights":"yes","params":{"text":"5B advertised; 5.334B safetensors parameters","source":"Tencent model card and HF safetensors metadata"},"facts":[{"label":"Inputs and output","value":"Document/image, audio and video inputs produce structured text/JSON with spatial boxes, temporal segments, OCR/ASR, captions and reports under one OmniSchema envelope"},{"label":"Architecture","value":"One shared perception encoder with thin modality-specific stems and merger/projector heads; audio/video alignment occurs inside perception; Youtu-LLM text decoder"},{"label":"License","value":"Custom License Terms of Youtu-Parsing (HF tag other, license_name youtu-parsing). Includes an overriding Additional Territorial Limitation: NOT INTENDED FOR USE WITHIN THE EUROPEAN UNION. Not unmodified Apache-2.0; third-party components retain their own terms"},{"label":"Serving","value":"Self-hosted vLLM plugin or Transformers/custom code; separate environments required by pinned Transformers versions; ffmpeg needed for audio/video. Examples are not evidence of a Tencent-hosted API"},{"label":"Limitations","value":"Autoregressive end-to-end decoding does not implement the parallel acceleration of older Youtu-Parsing. Lab reports remaining weaknesses in ASR, knowledge recognition, counting and geometric relationships"},{"label":"Related","value":"Separate from older 2.5B Youtu-Parsing, Youtu-VL, Youtu-LLM and Logics-Parsing-Omni"},{"label":"Date evidence","value":"Public HF conversion discussion reports attempts on this exact model on Oct 10 ET (Oct 11 00:37 UTC). This establishes public access before the Oct 29 paper; Oct 9 repo creation/upload alone does not establish public availability"}],"claims":["Lab claim: OmniDocBench v1.6 Overall 96.96 and OmniParsingBench average 75.08. The latter uses the lab's own re-scoring toolkit with schema repair and corrected annotations, excluding unavailable/problematic samples; not directly comparable with benchmark-author scores."],"sources":[{"kind":"repo","title":"Tencent Youtu-Parsing-Omni model card and weights","url":"https://huggingface.co/tencent/Youtu-Parsing-Omni"},{"kind":"repo","title":"Public conversion discussion: historical access evidence only","url":"https://huggingface.co/tencent/Youtu-Parsing-Omni/discussions/1","date":"2026-10-10"},{"kind":"repo","title":"Youtu-Parsing custom weight/code license","url":"https://huggingface.co/tencent/Youtu-Parsing-Omni/blob/main/LICENSE"},{"kind":"lab","title":"Tencent Youtu-Parsing-Omni technical report v1","url":"https://arxiv.org/abs/2610.39301v1"},{"kind":"repo","title":"Tencent official parsing code","url":"https://github.com/TencentCloudADP/youtu-parsing"}],"firstBenchmark":null},"organization":{"id":"tencent","name":"Tencent","slug":"tencent","country":null,"url":null,"color":"#A39E8F","mark":"TE"},"price":null,"answer":{"word":"Pending benchmarks.","scope":"Announced Oct 10, 2026; no independent results yet.","lead":"As of Nov 22, 2026, Youtu-Parsing-Omni is pending benchmarks on isitsota: announced Oct 10, 2026, with no independent results on the boards we track yet.","confidence":"low"},"status":"PENDING_BENCHMARKS","verdicts":[]}