Open-source alternatives to Longscape

✨ Verification through Spatial Assertion (ViSA) ✨

M I N D : Benchmarking Memory Consistency and Action Control in World Models TL;DR: The first open-domain closed-loop revisited benchmark for evaluating memory consistency and action control in world models

Official implementation of the DayDreamerpaper algorithm in TensorFlow 2.

Deep Hierarchical Planning from Pixels

Status: Stable release

A reimplementation of DreamerV3paper, a scalable and general reinforcement learning algorithm that masters a wide range of applications with fixed hyperparameters.

Transformers are Sample-Efficient World Models Vincent Micheli\, Eloi Alonso\, François Fleuret \* Denotes equal contribution

MorphoSim: An Interactive, Controllable, and Editable Language-guided 4D World Simulator

📄 arXiv 🏠 Project Page 🤗 Huggingface

Paper · arXiv · Project Page · 5-min Video

This is the official implementation of DreamerPro: Reconstruction-Free Model-Based Reinforcement Learning with Prototypical Representations in TensorFlow 2. A re-implementation of Temporal Predictive Coding for Model-Based Planning in Latent Space is also included.

Official implementation of NavMorph: A Self-Evolving World Model for Vision-and-Language Navigation in Continuous Environments (ICCV'25).

Weilun Feng 1,2 , Guoxin Fan 1,2 , Haotong Qin *3 , Chuanguang Yang †1 , Mingqiang Wu 1,2 , Yuqi Li 4 , Xiangqi Li 1,2 , Zhulin An †1 , Libo Huang 1 , Dingrui Wang 5 , Longlong Liao 6 , Michele Magno 3 , Yongjun Xu 1

Gabrijel Boduljak | Yushi Lan | Christian Rupprecht | Andrea Vedaldi

Out of Sight but Not Out of Mind: Hybrid Memory for Dynamic Video World Models

Xianjin Wu 1 , Dingkang Liang 1† , Tianrui Feng 1 , Kui Xia 2 , Yumeng Zhang 2 , Xiaofan Li 2 , Xiao Tan 2 , Xiang Bai 1 1 Huazhong University of Science and Technology, 2 Baidu Inc., China， † Project Lead

This repository contains PyTorch implementation for 3DFlowAction: Learning Cross-Embodiment Manipulation from 3D Flow World Model

$\dagger$ Project leader

This is the official code repository for the paper "Language Agents Meet Causality -- Bridging LLMs and Causal World Models".

World models play a crucial role in understanding and predicting the dynamics of the world, which is essential for video generation. However, existing world models are confined to specific scenarios such as gaming or driving, limiting their ability to capture the complexity of general world…

This repository contains the code for the paper Learning to Model the World with Language. We introduce Dynalang, an agent that leverages diverse types of language to solve tasks by using language to predict the future via a multimodal world model.

Official code of the paper Simple, Good, Fast: Self-Supervised World Models Free of Baggage. Published as a conference paper at ICLR 2025.

Currently WIP.

Pytorch version of Dreamer, which follows the original TF v2 codes (https://github.com/danijar/dreamerv2/tree/faf9e4c606e735f32c8b971a3092877265e49cc2).

Paper | Models (Coming Soon)