[go: up one dir, main page]

https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5.

\n","updatedAt":"2025-11-25T04:10:04.953Z","author":{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","fullname":"taesiri","name":"taesiri","type":"user","isPro":true,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":211,"isUserFollowing":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.8799499273300171},"editors":["taesiri"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg"],"reactions":[{"reaction":"πŸ”₯","users":["AdinaY","kabachuha","Jefxiong"],"count":3}],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2511.18870","authors":[{"_id":"69252c0d16eb3a9f131039d6","name":"Bing Wu","hidden":false},{"_id":"69252c0d16eb3a9f131039d7","name":"Chang Zou","hidden":false},{"_id":"69252c0d16eb3a9f131039d8","name":"Changlin Li","hidden":false},{"_id":"69252c0d16eb3a9f131039d9","name":"Duojun Huang","hidden":false},{"_id":"69252c0d16eb3a9f131039da","name":"Fang Yang","hidden":false},{"_id":"69252c0d16eb3a9f131039db","name":"Hao Tan","hidden":false},{"_id":"69252c0d16eb3a9f131039dc","name":"Jack Peng","hidden":false},{"_id":"69252c0d16eb3a9f131039dd","name":"Jianbing Wu","hidden":false},{"_id":"69252c0d16eb3a9f131039de","user":{"_id":"62a04f5e445b33e7d15e05ee","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/vS1YjcyOucCgFrde0lAvu.png","isPro":false,"fullname":"Xiong","user":"Jefxiong","type":"user"},"name":"Jiangfeng Xiong","status":"claimed_verified","statusLastChangedAt":"2025-11-26T09:28:15.062Z","hidden":false},{"_id":"69252c0d16eb3a9f131039df","name":"Jie Jiang","hidden":false},{"_id":"69252c0d16eb3a9f131039e0","name":"Linus","hidden":false},{"_id":"69252c0d16eb3a9f131039e1","name":"Patrol","hidden":false},{"_id":"69252c0d16eb3a9f131039e2","user":{"_id":"670bc2aac9c7e01d3db61243","avatarUrl":"/avatars/2bd5fba261e710a4c287fcf2f78b584f.svg","isPro":false,"fullname":"Peizhen Zhang","user":"maxpzzhang","type":"user"},"name":"Peizhen Zhang","status":"admin_assigned","statusLastChangedAt":"2025-11-25T12:19:00.711Z","hidden":false},{"_id":"69252c0d16eb3a9f131039e3","name":"Peng Chen","hidden":false},{"_id":"69252c0d16eb3a9f131039e4","name":"Penghao Zhao","hidden":false},{"_id":"69252c0d16eb3a9f131039e5","name":"Qi Tian","hidden":false},{"_id":"69252c0d16eb3a9f131039e6","name":"Songtao Liu","hidden":false},{"_id":"69252c0d16eb3a9f131039e7","user":{"_id":"67f7ab7ef241f0b92ce4b087","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/vDhBPKqfWjMI65K-kL2tm.jpeg","isPro":false,"fullname":"Weijie Kong","user":"Knightbreeze","type":"user"},"name":"Weijie Kong","status":"admin_assigned","statusLastChangedAt":"2025-11-25T12:19:44.134Z","hidden":false},{"_id":"69252c0d16eb3a9f131039e8","name":"Weiyan Wang","hidden":false},{"_id":"69252c0d16eb3a9f131039e9","name":"Xiao He","hidden":false},{"_id":"69252c0d16eb3a9f131039ea","name":"Xin Li","hidden":false},{"_id":"69252c0d16eb3a9f131039eb","name":"Xinchi Deng","hidden":false},{"_id":"69252c0d16eb3a9f131039ec","name":"Xuefei Zhe","hidden":false},{"_id":"69252c0d16eb3a9f131039ed","name":"Yang Li","hidden":false},{"_id":"69252c0d16eb3a9f131039ee","name":"Yanxin Long","hidden":false},{"_id":"69252c0d16eb3a9f131039ef","name":"Yuanbo Peng","hidden":false},{"_id":"69252c0d16eb3a9f131039f0","name":"Yue Wu","hidden":false},{"_id":"69252c0d16eb3a9f131039f1","name":"Yuhong Liu","hidden":false},{"_id":"69252c0d16eb3a9f131039f2","name":"Zhenyu Wang","hidden":false},{"_id":"69252c0d16eb3a9f131039f3","user":{"_id":"6316fe2a6d322c52490f90e8","avatarUrl":"/avatars/8c896b3733e183657009a426834a9dab.svg","isPro":false,"fullname":"Dai Zuozhuo","user":"zyand","type":"user"},"name":"Zuozhuo Dai","status":"admin_assigned","statusLastChangedAt":"2025-11-25T12:19:54.939Z","hidden":false},{"_id":"69252c0d16eb3a9f131039f4","name":"Bo Peng","hidden":false},{"_id":"69252c0d16eb3a9f131039f5","name":"Coopers Li","hidden":false},{"_id":"69252c0d16eb3a9f131039f6","name":"Gu Gong","hidden":false},{"_id":"69252c0d16eb3a9f131039f7","name":"Guojian Xiao","hidden":false},{"_id":"69252c0d16eb3a9f131039f8","name":"Jiahe Tian","hidden":false},{"_id":"69252c0d16eb3a9f131039f9","name":"Jiaxin Lin","hidden":false},{"_id":"69252c0d16eb3a9f131039fa","name":"Jie Liu","hidden":false},{"_id":"69252c0d16eb3a9f131039fb","name":"Jihong Zhang","hidden":false},{"_id":"69252c0d16eb3a9f131039fc","name":"Jiesong Lian","hidden":false},{"_id":"69252c0d16eb3a9f131039fd","name":"Kaihang Pan","hidden":false},{"_id":"69252c0d16eb3a9f131039fe","name":"Lei Wang","hidden":false},{"_id":"69252c0d16eb3a9f131039ff","name":"Lin Niu","hidden":false},{"_id":"69252c0d16eb3a9f13103a00","name":"Mingtao Chen","hidden":false},{"_id":"69252c0d16eb3a9f13103a01","name":"Mingyang Chen","hidden":false},{"_id":"69252c0d16eb3a9f13103a02","name":"Mingzhe Zheng","hidden":false},{"_id":"69252c0d16eb3a9f13103a03","name":"Miles Yang","hidden":false},{"_id":"69252c0d16eb3a9f13103a04","name":"Qiangqiang Hu","hidden":false},{"_id":"69252c0d16eb3a9f13103a05","name":"Qi Yang","hidden":false},{"_id":"69252c0d16eb3a9f13103a06","name":"Qiuyong Xiao","hidden":false},{"_id":"69252c0d16eb3a9f13103a07","name":"Runzhou Wu","hidden":false},{"_id":"69252c0d16eb3a9f13103a08","name":"Ryan Xu","hidden":false},{"_id":"69252c0d16eb3a9f13103a09","name":"Rui Yuan","hidden":false},{"_id":"69252c0d16eb3a9f13103a0a","name":"Shanshan Sang","hidden":false},{"_id":"69252c0d16eb3a9f13103a0b","name":"Shisheng Huang","hidden":false},{"_id":"69252c0d16eb3a9f13103a0c","name":"Siruis Gong","hidden":false},{"_id":"69252c0d16eb3a9f13103a0d","name":"Shuo Huang","hidden":false},{"_id":"69252c0d16eb3a9f13103a0e","name":"Weiting Guo","hidden":false},{"_id":"69252c0d16eb3a9f13103a0f","name":"Xiang Yuan","hidden":false},{"_id":"69252c0d16eb3a9f13103a10","name":"Xiaojia Chen","hidden":false},{"_id":"69252c0d16eb3a9f13103a11","name":"Xiawei Hu","hidden":false},{"_id":"69252c0d16eb3a9f13103a12","name":"Wenzhi Sun","hidden":false},{"_id":"69252c0d16eb3a9f13103a13","name":"Xiele Wu","hidden":false},{"_id":"69252c0d16eb3a9f13103a14","name":"Xianshun Ren","hidden":false},{"_id":"69252c0d16eb3a9f13103a15","name":"Xiaoyan Yuan","hidden":false},{"_id":"69252c0d16eb3a9f13103a16","user":{"_id":"66f6627e08be8ab9ab0833f2","avatarUrl":"/avatars/871a0862c22e0e4f8829144953f81d85.svg","isPro":false,"fullname":"mi","user":"mxy123","type":"user"},"name":"Xiaoyue Mi","status":"claimed_verified","statusLastChangedAt":"2025-11-28T13:49:04.776Z","hidden":false},{"_id":"69252c0d16eb3a9f13103a17","name":"Yepeng Zhang","hidden":false},{"_id":"69252c0d16eb3a9f13103a18","name":"Yifu Sun","hidden":false},{"_id":"69252c0d16eb3a9f13103a19","name":"Yiting Lu","hidden":false},{"_id":"69252c0d16eb3a9f13103a1a","name":"Yitong Li","hidden":false},{"_id":"69252c0d16eb3a9f13103a1b","name":"You Huang","hidden":false},{"_id":"69252c0d16eb3a9f13103a1c","name":"Yu Tang","hidden":false},{"_id":"69252c0d16eb3a9f13103a1d","name":"Yixuan Li","hidden":false},{"_id":"69252c0d16eb3a9f13103a1e","name":"Yuhang Deng","hidden":false},{"_id":"69252c0d16eb3a9f13103a1f","name":"Yuan Zhou","hidden":false},{"_id":"69252c0d16eb3a9f13103a20","name":"Zhichao Hu","hidden":false},{"_id":"69252c0d16eb3a9f13103a21","user":{"_id":"64556769fbe00f9e73c1bddb","avatarUrl":"/avatars/dc0a2d9cf91190d722cb6613072a6556.svg","isPro":false,"fullname":"ZHIGUANG LIU","user":"lz7fdmu","type":"user"},"name":"Zhiguang Liu","status":"admin_assigned","statusLastChangedAt":"2025-11-25T12:23:02.416Z","hidden":false},{"_id":"69252c0d16eb3a9f13103a22","name":"Zhihe Yang","hidden":false},{"_id":"69252c0d16eb3a9f13103a23","name":"Zilin Yang","hidden":false},{"_id":"69252c0d16eb3a9f13103a24","name":"Zhenzhi Lu","hidden":false},{"_id":"69252c0d16eb3a9f13103a25","name":"Zixiang Zhou","hidden":false},{"_id":"69252c0d16eb3a9f13103a26","name":"Zhao Zhong","hidden":false}],"publishedAt":"2025-11-24T08:22:07.000Z","submittedOnDailyAt":"2025-11-25T01:39:58.047Z","title":"HunyuanVideo 1.5 Technical Report","submittedOnDailyBy":{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","isPro":true,"fullname":"taesiri","user":"taesiri","type":"user"},"summary":"We present HunyuanVideo 1.5, a lightweight yet powerful open-source video generation model that achieves state-of-the-art visual quality and motion coherence with only 8.3 billion parameters, enabling efficient inference on consumer-grade GPUs. This achievement is built upon several key components, including meticulous data curation, an advanced DiT architecture featuring selective and sliding tile attention (SSTA), enhanced bilingual understanding through glyph-aware text encoding, progressive pre-training and post-training, and an efficient video super-resolution network. Leveraging these designs, we developed a unified framework capable of high-quality text-to-video and image-to-video generation across multiple durations and resolutions.Extensive experiments demonstrate that this compact and proficient model establishes a new state-of-the-art among open-source video generation models. By releasing the code and model weights, we provide the community with a high-performance foundation that lowers the barrier to video creation and research, making advanced video generation accessible to a broader audience. All open-source assets are publicly available at https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5.","upvotes":25,"discussionId":"69252c0d16eb3a9f13103a27","githubRepo":"https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5","githubRepoAddedBy":"auto","ai_summary":"HunyuanVideo 1.5 is a lightweight video generation model with state-of-the-art visual quality and motion coherence, using a DiT architecture with SSTA and an efficient video super-resolution network.","ai_keywords":["DiT architecture","selective and sliding tile attention","SSTA","glyph-aware text encoding","progressive pre-training","post-training","video super-resolution network","text-to-video","image-to-video"],"githubStars":3562},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","isPro":true,"fullname":"taesiri","user":"taesiri","type":"user"},{"_id":"67bbade8a8c89b98ec377944","avatarUrl":"/avatars/640803ef641decd5c30894155bf09b6a.svg","isPro":false,"fullname":"Urodoc Oncall","user":"UDCAI","type":"user"},{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"},{"_id":"645e553c3b6d85c65e8b0e54","avatarUrl":"/avatars/1fffc6499b9d65b21a895ca96f03b781.svg","isPro":false,"fullname":"Steven","user":"yijunyang","type":"user"},{"_id":"61af81009f77f7b669578f95","avatarUrl":"/avatars/fb50773ac49948940eb231834ee6f2fd.svg","isPro":false,"fullname":"rotem israeli","user":"irotem98","type":"user"},{"_id":"63c5d43ae2804cb2407e4d43","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1673909278097-noauth.png","isPro":false,"fullname":"xziayro","user":"xziayro","type":"user"},{"_id":"66fb03d6b505f1a04c39d935","avatarUrl":"/avatars/e9b830c460ec02037758c9b3469bb8ad.svg","isPro":false,"fullname":"Xuanlang Dai","user":"VVitaminD","type":"user"},{"_id":"677272184d148b904333e874","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/5dUau7gxLk4Wm1TiiJJri.jpeg","isPro":false,"fullname":"Efstathios Karypidis","user":"Sta8is","type":"user"},{"_id":"63a369d98c0c89dcae3b8329","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63a369d98c0c89dcae3b8329/AiH2zjy1cnt9OADAAZMLD.jpeg","isPro":false,"fullname":"Adina Yakefu","user":"AdinaY","type":"user"},{"_id":"63177d85f957903db971a173","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1665094764329-63177d85f957903db971a173.png","isPro":false,"fullname":"Artem","user":"kabachuha","type":"user"},{"_id":"66313d0ac57e46020d0421b6","avatarUrl":"/avatars/d0c75e5a4fef61b37632892632565075.svg","isPro":false,"fullname":"Yannay Alon","user":"YannayAlon","type":"user"},{"_id":"63c6cb6a50cc81901da65e15","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63c6cb6a50cc81901da65e15/t4LN1BPCFlwbSJ9GD9YDd.jpeg","isPro":true,"fullname":"ThΓ©o Pomies","user":"theopomies","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">
Papers
arxiv:2511.18870

HunyuanVideo 1.5 Technical Report

Published on Nov 24, 2025
Β· Submitted by
taesiri
on Nov 25, 2025
Authors:
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,

Abstract

HunyuanVideo 1.5 is a lightweight video generation model with state-of-the-art visual quality and motion coherence, using a DiT architecture with SSTA and an efficient video super-resolution network.

AI-generated summary

We present HunyuanVideo 1.5, a lightweight yet powerful open-source video generation model that achieves state-of-the-art visual quality and motion coherence with only 8.3 billion parameters, enabling efficient inference on consumer-grade GPUs. This achievement is built upon several key components, including meticulous data curation, an advanced DiT architecture featuring selective and sliding tile attention (SSTA), enhanced bilingual understanding through glyph-aware text encoding, progressive pre-training and post-training, and an efficient video super-resolution network. Leveraging these designs, we developed a unified framework capable of high-quality text-to-video and image-to-video generation across multiple durations and resolutions.Extensive experiments demonstrate that this compact and proficient model establishes a new state-of-the-art among open-source video generation models. By releasing the code and model weights, we provide the community with a high-performance foundation that lowers the barrier to video creation and research, making advanced video generation accessible to a broader audience. All open-source assets are publicly available at https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5.

Community

Paper submitter
β€’
edited Nov 25, 2025

We present HunyuanVideo 1.5, a lightweight yet powerful open-source video generation model that achieves state-of-the-art visual quality and motion coherence with only 8.3 billion parameters, enabling efficient inference on consumer-grade GPUs. This achievement is built upon several key components, including meticulous data curation, an advanced DiT architecture featuring selective and sliding tile attention (SSTA), enhanced bilingual understanding through glyph-aware text encoding, progressive pre-training and post-training, and an efficient video super-resolution network. Leveraging these designs, we developed a unified framework capable of high-quality text-to-video and image-to-video generation across multiple durations and resolutions.Extensive experiments demonstrate that this compact and proficient model establishes a new state-of-the-art among open-source video generation models. By releasing the code and model weights, we provide the community with a high-performance foundation that lowers the barrier to video creation and research, making advanced video generation accessible to a broader audience. All open-source assets are publicly available at https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5.

Sign up or log in to comment

Models citing this paper 2

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2511.18870 in a dataset README.md to link it from this page.

Spaces citing this paper 104

Collections including this paper 4