LongCat-Video Technical Report

arXiv 2025 25.3 method, system

TLDR

LongCat-Video is a 13.6B parameter video generation model excelling in long video generation, aiming toward world models.

Reasoning

Strengths: strong performance, efficient long video generation, unified architecture. Weaknesses: limited novelty as a step toward world models; no direct world model capabilities demonstrated.

Read-first score

Read-first score 25.3, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 6.

Recency 8%

86.7

Uses a gentle age decay so recent papers surface without erasing older foundations. 2025

Reproducibility 25%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=code

Methodology quality 25%

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=none

Topical relevance 42%

8.6

Uses existing LLM keyword relevance scores normalized to 0-100. world model,world simulator,generative world model,interactive world model,video world model,world dynamics prediction,model-based reinforcement learning world model

Field roles

Frontier

Rank sensitivity

Stability: volatile; rank range: 46.

Keyword Scores

world model

video world model

world simulator

generative world model

interactive world model

world dynamics prediction

model-based reinforcement learning world model

Resources

Paper