🚀 Thrilled to introduce Vidu S2: real-time AI video you can talk to, direct, and real-time editing.</p>\n<p>🎙️ Vidu S2-Avatar<br>Interactive characters at 720p and 25–42 FPS, with stronger instruction following, expressive full-body motion, and dancing. Introduce new reference images anytime to change outfits, interact with objects, or switch scenes mid-conversation.</p>\n<p>🎨 Vidu S2-Editing<br>Transform an incoming video stream in real time while preserving motion and timing: style transfer, virtual try-on, character replacement, and background replacement.</p>\n<p>🥽 Spatial video<br>Bring generated characters and live video edits into stereoscopic video for immersive VR experiences.</p>\n<p>🏆 SOTA results across all five public benchmarks in our evaluation, spanning avatar generation and video editing.</p>\n<p>🎮 Try it: <a href=\"https://vidu.com/vidu-stream\" rel=\"nofollow\">https://vidu.com/vidu-stream</a><br>🛠️ API: <a href=\"http://platform.vidu.com/vidu-stream/doc\" rel=\"nofollow\">http://platform.vidu.com/vidu-stream/doc</a><br>📄 Technical report: <a href=\"https://arxiv.org/pdf/2609.11638\" rel=\"nofollow\">https://arxiv.org/pdf/2609.11638</a></p>\n","updatedAt":"2026-09-15T02:37:41.834Z","author":{"_id":"66c0a08bac74db25de8427ec","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/66c0a08bac74db25de8427ec/9D-piDBZqSt6KNkHImmkv.jpeg","fullname":"Jintao Zhang","name":"jt-zhang","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":61,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8098416328430176},"editors":["jt-zhang"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/66c0a08bac74db25de8427ec/9D-piDBZqSt6KNkHImmkv.jpeg"],"reactions":[{"reaction":"❤️","users":["jt-zhang","timchowwww","SempraETY"],"count":3}],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2609.11638","authors":[{"_id":"6aa61fef7ba345d44ad146e9","name":"Jintao Zhang","hidden":false},{"_id":"6aa61fef7ba345d44ad146ea","name":"Kai Jiang","hidden":false},{"_id":"6aa61fef7ba345d44ad146eb","name":"Jintao Chen","hidden":false},{"_id":"6aa61fef7ba345d44ad146ec","name":"Xu Wang","hidden":false},{"_id":"6aa61fef7ba345d44ad146ed","name":"Deyuan Liu","hidden":false},{"_id":"6aa61fef7ba345d44ad146ee","name":"Jungang Li","hidden":false},{"_id":"6aa61fef7ba345d44ad146ef","name":"Dechuang Chen","hidden":false},{"_id":"6aa61fef7ba345d44ad146f0","name":"Ming Lin","hidden":false},{"_id":"6aa61fef7ba345d44ad146f1","name":"Jingjiang Zhou","hidden":false},{"_id":"6aa61fef7ba345d44ad146f2","name":"Haopeng Jin","hidden":false},{"_id":"6aa61fef7ba345d44ad146f3","name":"Qi Jia","hidden":false},{"_id":"6aa61fef7ba345d44ad146f4","name":"Xiaohang Wang","hidden":false},{"_id":"6aa61fef7ba345d44ad146f5","name":"Yaole Wang","hidden":false},{"_id":"6aa61fef7ba345d44ad146f6","name":"Zhanqiang Zhang","hidden":false},{"_id":"6aa61fef7ba345d44ad146f7","name":"Ran Li","hidden":false},{"_id":"6aa61fef7ba345d44ad146f8","name":"Zhengkun Huang","hidden":false},{"_id":"6aa61fef7ba345d44ad146f9","name":"Shuyue Xiong","hidden":false},{"_id":"6aa61fef7ba345d44ad146fa","name":"Yuji Wang","hidden":false},{"_id":"6aa61fef7ba345d44ad146fb","name":"Zikun Dai","hidden":false},{"_id":"6aa61fef7ba345d44ad146fc","name":"Hui He","hidden":false},{"_id":"6aa61fef7ba345d44ad146fd","name":"Yang Luo","hidden":false},{"_id":"6aa61fef7ba345d44ad146fe","name":"Mang Ning","hidden":false},{"_id":"6aa61fef7ba345d44ad146ff","name":"Weiqi Feng","hidden":false},{"_id":"6aa61fef7ba345d44ad14700","name":"Chengyang Ye","hidden":false},{"_id":"6aa61fef7ba345d44ad14701","name":"Xinyue Lin","hidden":false},{"_id":"6aa61fef7ba345d44ad14702","name":"Min Zhao","hidden":false},{"_id":"6aa61fef7ba345d44ad14703","name":"Hongzhou Zhu","hidden":false},{"_id":"6aa61fef7ba345d44ad14704","name":"Hengkai Tan","hidden":false},{"_id":"6aa61fef7ba345d44ad14705","name":"Zeyuan Wang","hidden":false},{"_id":"6aa61fef7ba345d44ad14706","name":"Chendong Xiang","hidden":false},{"_id":"6aa61fef7ba345d44ad14707","name":"Kaiwen Zheng","hidden":false},{"_id":"6aa61fef7ba345d44ad14708","name":"Zhijie Deng","hidden":false},{"_id":"6aa61fef7ba345d44ad14709","name":"Fan Bao","hidden":false},{"_id":"6aa61fef7ba345d44ad1470a","name":"Jianfei Chen","hidden":false},{"_id":"6aa61fef7ba345d44ad1470b","name":"Jun Zhu","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/66c0a08bac74db25de8427ec/b5nSODVi3mSrwIzgvQosn.png"],"publishedAt":"2026-09-10T00:00:00.000Z","submittedOnDailyAt":"2026-09-15T00:00:00.000Z","title":"Vidu S2: Real-Time Interactive, Editable, and Spatial Video Generation","submittedOnDailyBy":{"_id":"66c0a08bac74db25de8427ec","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/66c0a08bac74db25de8427ec/9D-piDBZqSt6KNkHImmkv.jpeg","isPro":false,"fullname":"Jintao Zhang","user":"jt-zhang","type":"user","name":"jt-zhang"},"summary":"We present Vidu S2, which comprises Vidu S2-Avatar, a real-time interactive digital-character model, and Vidu S2-Editing, a real-time video editing model. Moreover, we explore the feasibility of real-time spatial video generation for both Vidu S2-Avatar and Vidu S2-Editing. Compared with Vidu S1, Vidu S2-Avatar supports real-time 720p video generation, generation with dynamic references that can be updated at any moment, and stronger instruction following, such as dancing. Vidu S2-Editing supports editing a video stream in real time, including style rendering, clothing replacement, character replacement, and background replacement. Experiments show that Vidu S2 outperforms all baselines. A playable online demo is available at https://vidu.com/vidu-stream.","upvotes":127,"discussionId":"6aa61ff07ba345d44ad1470c","projectPage":"https://www.vidu.com/vidu-stream","githubRepo":"https://github.com/shengshu-ai/Vidu-S","githubRepoAddedBy":"user","ai_summary":"Vidu S2 introduces real-time interactive avatar and video editing models that support high-resolution spatial video generation and dynamic reference updates.","ai_keywords":["real-time video generation","spatial video generation","dynamic references","instruction following","video editing","style rendering","character replacement","background replacement"],"ai_summary_model":"thinkingmachines/Inkling-Small","githubStars":253,"organization":{"_id":"628735cbc83a2d6ab8d14a66","name":"Tsinghua","fullname":"Tsinghua University","avatar":"https://www.gravatar.com/avatar/6c5c1441e3283e7543342e59277ea219?d=retro&size=100"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"66c0a08bac74db25de8427ec","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/66c0a08bac74db25de8427ec/9D-piDBZqSt6KNkHImmkv.jpeg","isPro":false,"fullname":"Jintao Zhang","user":"jt-zhang","type":"user"},{"_id":"64b8f7e10d8834cb3241beb5","avatarUrl":"/avatars/eb9c76b5539a4f8899dc9f43731e59a6.svg","isPro":false,"fullname":"liran","user":"liran12345","type":"user"},{"_id":"69a655c674b0a43e9c2239dd","avatarUrl":"/avatars/95d1fd7724f25f6009b2ed246f6d64b5.svg","isPro":false,"fullname":"zhangzhanqiang","user":"zhangzhanqiang","type":"user"},{"_id":"67ff605e58ed263257c3d896","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/adCF359QLIcGm6G4XpiuX.jpeg","isPro":false,"fullname":"Jake Dai","user":"Clare0354","type":"user"},{"_id":"6698a1efbf3edea3110a6e66","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6698a1efbf3edea3110a6e66/WzwIu_8NKEcN10CEw8G8O.jpeg","isPro":false,"fullname":"Weiqi Feng","user":"Fengwq22","type":"user"},{"_id":"673178311b391b5e0ec35133","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/3q_FcTTZGB1PdO_a4MGbG.png","isPro":false,"fullname":"Hengkai Tan","user":"bernardtan","type":"user"},{"_id":"6912e9c6fab44759e4ede84e","avatarUrl":"/avatars/63460e9d101bb71bb398f1281235ee45.svg","isPro":false,"fullname":"LiuJiabao","user":"LiuJiabao","type":"user"},{"_id":"6949058a4cb4a06b008088a9","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/RaWhJSj3J1WZWF02Treok.png","isPro":false,"fullname":"Mengchen Cai","user":"Yourchen","type":"user"},{"_id":"6a01704890b6a89522fa2e42","avatarUrl":"/avatars/68622b66e02218f6cbf97e8cfe289878.svg","isPro":false,"fullname":"Fan Bao","user":"fanbao123","type":"user"},{"_id":"665ebffc5baf7febc729fdd7","avatarUrl":"/avatars/aae8c6be2c778288a3e94708f1af19e5.svg","isPro":false,"fullname":"wangtian shen","user":"swt20010801","type":"user"},{"_id":"6a2683029a4a7577ed195e35","avatarUrl":"/avatars/896db140be1ef2dc78b2168c0ca7ca09.svg","isPro":false,"fullname":"huangshuhe","user":"shuhehuang","type":"user"},{"_id":"654ca1b26149b7666ad47497","avatarUrl":"/avatars/2243c73a74e3e839749e42943edf640d.svg","isPro":false,"fullname":"Hanyangyu","user":"Hanyangyu","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":3,"organization":{"_id":"628735cbc83a2d6ab8d14a66","name":"Tsinghua","fullname":"Tsinghua University","avatar":"https://www.gravatar.com/avatar/6c5c1441e3283e7543342e59277ea219?d=retro&size=100"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2609/2609.11638.md","query":{}}">
Vidu S2: Real-Time Interactive, Editable, and Spatial Video Generation
Abstract
Vidu S2 introduces real-time interactive avatar and video editing models that support high-resolution spatial video generation and dynamic reference updates.
We present Vidu S2, which comprises Vidu S2-Avatar, a real-time interactive digital-character model, and Vidu S2-Editing, a real-time video editing model. Moreover, we explore the feasibility of real-time spatial video generation for both Vidu S2-Avatar and Vidu S2-Editing. Compared with Vidu S1, Vidu S2-Avatar supports real-time 720p video generation, generation with dynamic references that can be updated at any moment, and stronger instruction following, such as dancing. Vidu S2-Editing supports editing a video stream in real time, including style rendering, clothing replacement, character replacement, and background replacement. Experiments show that Vidu S2 outperforms all baselines. A playable online demo is available at https://vidu.com/vidu-stream.
Community
🚀 Thrilled to introduce Vidu S2: real-time AI video you can talk to, direct, and real-time editing.
🎙️ Vidu S2-Avatar
Interactive characters at 720p and 25–42 FPS, with stronger instruction following, expressive full-body motion, and dancing. Introduce new reference images anytime to change outfits, interact with objects, or switch scenes mid-conversation.
🎨 Vidu S2-Editing
Transform an incoming video stream in real time while preserving motion and timing: style transfer, virtual try-on, character replacement, and background replacement.
🥽 Spatial video
Bring generated characters and live video edits into stereoscopic video for immersive VR experiences.
🏆 SOTA results across all five public benchmarks in our evaluation, spanning avatar generation and video editing.
🎮 Try it: https://vidu.com/vidu-stream
🛠️ API: http://platform.vidu.com/vidu-stream/doc
📄 Technical report: https://arxiv.org/pdf/2609.11638
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2609.11638 in a model README.md to link it from this page.
Cite arxiv.org/abs/2609.11638 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2609.11638 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.