LongCat-Video Technical Report
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915581583687680 |
|---|---|
| author | Meituan LongCat Team Cai, Xunliang Huang, Qilong Kang, Zhuoliang Li, Hongyu Liang, Shijun Ma, Liya Ren, Siyu Wei, Xiaoming Xie, Rixu Zhang, Tong |
| author_facet | Meituan LongCat Team Cai, Xunliang Huang, Qilong Kang, Zhuoliang Li, Hongyu Liang, Shijun Ma, Liya Ren, Siyu Wei, Xiaoming Xie, Rixu Zhang, Tong |
| contents | Video generation is a critical pathway toward world models, with efficient long video inference as a key capability. Toward this end, we introduce LongCat-Video, a foundational video generation model with 13.6B parameters, delivering strong performance across multiple video generation tasks. It particularly excels in efficient and high-quality long video generation, representing our first step toward world models. Key features include: Unified architecture for multiple tasks: Built on the Diffusion Transformer (DiT) framework, LongCat-Video supports Text-to-Video, Image-to-Video, and Video-Continuation tasks with a single model; Long video generation: Pretraining on Video-Continuation tasks enables LongCat-Video to maintain high quality and temporal coherence in the generation of minutes-long videos; Efficient inference: LongCat-Video generates 720p, 30fps videos within minutes by employing a coarse-to-fine generation strategy along both the temporal and spatial axes. Block Sparse Attention further enhances efficiency, particularly at high resolutions; Strong performance with multi-reward RLHF: Multi-reward RLHF training enables LongCat-Video to achieve performance on par with the latest closed-source and leading open-source models. Code and model weights are publicly available to accelerate progress in the field. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_22200 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LongCat-Video Technical Report Meituan LongCat Team Cai, Xunliang Huang, Qilong Kang, Zhuoliang Li, Hongyu Liang, Shijun Ma, Liya Ren, Siyu Wei, Xiaoming Xie, Rixu Zhang, Tong Computer Vision and Pattern Recognition Video generation is a critical pathway toward world models, with efficient long video inference as a key capability. Toward this end, we introduce LongCat-Video, a foundational video generation model with 13.6B parameters, delivering strong performance across multiple video generation tasks. It particularly excels in efficient and high-quality long video generation, representing our first step toward world models. Key features include: Unified architecture for multiple tasks: Built on the Diffusion Transformer (DiT) framework, LongCat-Video supports Text-to-Video, Image-to-Video, and Video-Continuation tasks with a single model; Long video generation: Pretraining on Video-Continuation tasks enables LongCat-Video to maintain high quality and temporal coherence in the generation of minutes-long videos; Efficient inference: LongCat-Video generates 720p, 30fps videos within minutes by employing a coarse-to-fine generation strategy along both the temporal and spatial axes. Block Sparse Attention further enhances efficiency, particularly at high resolutions; Strong performance with multi-reward RLHF: Multi-reward RLHF training enables LongCat-Video to achieve performance on par with the latest closed-source and leading open-source models. Code and model weights are publicly available to accelerate progress in the field. |
| title | LongCat-Video Technical Report |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.22200 |