Official Implementation of "Meta Flow Maps enable scalable reward alignment"

Official implementation of Decoupled MeanFlow

Advancing Open-source World Models

Enjoy the magic of Diffusion models!

Official PyTorch Implementation of "Flow Map Distillation Without Data"

Inferix: A Block-Diffusion based Next-Generation Inference Engine for World Simulation

Pytorch implementation of MeanFlow on ImageNet and CIFAR10

Terminal Velocity Matching

SAM 3D Objects

Chicago undergraduate mathematics bibliography

Material for gpu-mode lectures

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

Tiny AutoEncoder for Hunyuan Video (and other video models)

A Curated List of Awesome Works in World Modeling, Aiming to Serve as a One-stop Resource for Researchers, Practitioners, and Enthusiasts Interested in World Modeling.

Optimal transport tools implemented with the JAX framework, to solve large scale matching problems of any flavor.

Krea Realtime 14B. An open-source realtime AI video model.

[ICLR 2026] rCM: SOTA JVP-Based Diffusion Distillation & Few-Step Video Generation & Scaling Up sCM/MeanFlow

[ICLR 2026] Official Repo for Rolling Forcing: Autoregressive Long Video Diffusion in Real Time

Hunyuan3D-Omni: A Unified Framework for Controllable Generation of 3D Assets

Aiming to integrate most existing feature caching-based diffusion acceleration schemes into a unified framework.

Light Image Video Generation Inference Framework

ReconViaGen: Towards Accurate Multi-view 3D Object Reconstruction via Generation

Make self forcing endless. Add cache purging. Add prompt controllability.

A beautiful, simple, clean, and responsive Jekyll theme for academics

A reimplementation of Stable Diffusion 3.5 in pure PyTorch

📹 A more flexible framework that can generate videos at any resolution and creates videos from images.

An Infinitely Large Napkin

Official implementation of "Force Prompting: Video Generation Models Can Learn and Generalize Physics-based Control Signals" (NeurIPS 2025)