LongCat-Video Technical Report

Papers with Code Trending 10/25/25, 07:41 AM Papers

Summary

LongCat-Video is a 13.6B parameter video generation model based on Diffusion Transformer, supporting text-to-video, image-to-video, and video-continuation tasks with efficient long video generation using coarse-to-fine and block sparse attention.

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.

Original Article

View Cached Full Text

Cached at: 05/23/26, 12:22 AM

Paper page - LongCat-Video Technical Report

Source: https://huggingface.co/papers/2510.22200 Published on Oct 25, 2025

Submitted byhttps://huggingface.co/taesiri

taesirion Oct 28, 2025

Authors:

Abstract

LongCat-Video, a 13.6B parameter video generation model based on the Diffusion Transformer framework, excels in efficient and high-quality long video generation across multiple tasks using unified architecture, coarse-to-fine generation, and block sparse attention.

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 theDiffusion Transformer(DiT) framework, LongCat-Video supportsText-to-Video,Image-to-Video, andVideo-Continuationtasks with a single model; Long video generation: Pretraining onVideo-Continuationtasks enables LongCat-Video to maintain high quality andtemporal coherencein the generation of minutes-long videos; Efficient inference: LongCat-Video generates 720p, 30fps videos within minutes by employing acoarse-to-fine generationstrategy along both the temporal and spatial axes.Block Sparse Attentionfurther enhances efficiency, particularly at high resolutions; Strong performance withmulti-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.

View arXiv page View PDF GitHub2.6k Add to collection

Get this paper in your agent:

hf papers read 2510\.22200

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash