[ICML'26] Code and website for Self-Flow: Self-Supervised Flow Matching for Scalable Multi-Modal Synthesis

Repo for SeedVR2 (ICLR2026) & SeedVR (CVPR2025 Highlight)

Matrix-Game 3.5: Enhancing Real-Time Streaming Interactive World Models with Patch Memory

On-Policy Self-Distillation for Post-Training Few-Step Autoregressive Video Generators

Code for MIRA: Multiplayer Interactive World Models with Representation Autoencoders

Infinite Worlds with Versatile Interactions

Full-stack open-source interactive long-horizon world model.

Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligence

Hy3 (295B A21B), a leading reasoning and agent model in its size, with great cost efficiency.

A Unified Multimodal Model for Scientific Image Understanding and Generation

Official code repository for the paper "Hallucination in World Models is Predictable and Preventable".

This project is the official implementation of "UnityShots: Memory-Driven Multi-Shot Audio-Video Generation with Boundary-Aware Gating"

WBench: A Comprehensive Multi-turn Benchmark for Interactive Video World Model Evaluation

Data-Forcing Distillation (DFD): restoring diversity and fidelity in few-step video generation — text-to-video (Wan2.1) & image-to-video (Cosmos), built on NVIDIA FastGen.

Qwen-AgentWorld: Language World Models for General Agents

Advancing Open-source World Models

agentic paper reviewer

AI PPT赛道终结者，史上最最最强 PPT Skill！！！ 使用GPT生成豪华的图片格式PPT，然后转换为完全可编辑的PPTX文件。

A programmable, explicit world model on Godot — worlds are pure JSON run by a fixed primitives + interpreter engine. Built by Claude, for Claude.

Official implementation of LoomVideo: Unifying Multimodal Inputs into Video Generation and Editing

[CVPR 2026 Oral] Official implementation for ChordEdit: One-Step Low-Energy Transport for Image Editing

[NeurIPS 2025] An official implementation of Flow-GRPO: Training Flow Matching Models via Online RL

Multimodal RL training framework for diffusion & omni models

[TOG 2024]StyleCrafter: Enhancing Stylized Text-to-Video Generation with Style Adapter

Implementation of Gamma-World: Generative Multi-Agent World Modeling Beyond Two Players

A Minimal and Elegant Framework & Tutorial for Real-Time Interactive World Models

Long-form audio-visual generation evaluation framework.