[ArXiv 25] Stable Video Infinity: Infinite-Length Video Generation with Error Recycling

Official PyTorch Implementation of "Diffusion Transformers with Representation Autoencoders"

A comprehensive list of papers for the definition of World Models and using World Models for General Video Generation, Embodied AI, and Autonomous Driving, including papers, codes, and related webs…

VideoNSA: Native Sparse Attention Scales Video Understanding

Official Repo for Self-Forcing++ High Quality Long Video Generation

LongLive: Real-time Interactive Long Video Generation

HunyuanImage-3.0: A Powerful Native Multimodal Model for Image Generation

Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

SpatialVID: A Large-Scale Video Dataset with Spatial Annotations

Tracking the latest and greatest research papers on video generation.

Unofficial extension implementation of Self-Forcing to support I2V && 14B training.

4-steps distilled version of Wan2.2-TI2V-5B

A collection of paper/projects that trains flow matching model/policies via RL.

Pusa: Thousands Timesteps Video Diffusion Model

Model analyzer in PyTorch

EasyR1: An Efficient, Scalable, Multi-Modality RL Training Framework based on veRL

A unified inference and post-training framework for accelerated video generation.

(CVPR 2025) From Slow Bidirectional to Fast Autoregressive Video Diffusion Models

code for "Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion"

Official codebase for "Self Forcing: Bridging Training and Inference in Autoregressive Video Diffusion" (NeurIPS 2025 Spotlight)

T2I-ReasonBench: Benchmarking Reasoning-Informed Text-to-Image Generation

SkyReels-V2: Infinite-length Film Generative model

Use Claude Code or Cursor CLI on mobile and web with Claude Code UI. Claude Code UI free open source webui/GUI that helps you manage your Claude Code session and projects remotely

An open-source AI agent that brings the power of Gemini directly into your terminal.

Matrix-Game 2.0: An Open-Source, Real-Time, and Streaming Interactive World Model

Reference PyTorch implementation and models for DINOv3

Stand-In is a lightweight, plug-and-play framework for identity-preserving video generation.

[Siggraph '23] NeRSemble: Neural Radiance Field Reconstruction of Human Heads