Global

LeRobot aims to provide models, datasets, and tools for real-world robotics in PyTorch. The goal is to lower the barrier to entry so that everyone can contribute to and benefit from shared datasets and pretrained models.

[Lumina具身智能社区] 具身智能技术指南 Embodied-AI-Guide

A framework for efficient model inference with omni-modality models

The scripts load VPP2 directly from this checkout; installing the vpp2 package is unnecessary. txt.

Framework · Domains · Results · Quick Start · Citation · 简体中文

OpenWAM is an open research stack for systematically developing World-Action Models (WAMs). It turns tightly coupled design choices into modular components and controlled experiments. It consists of:

openpi holds open-source models and packages for robotics, published by the Physical Intelligence team.

json") to use FlashAttention-3 or xformers once installed.

白箱AGI架构探索：元认知（自我认知循环）、持续学习（知识飞轮）、世界模型（条件空间+语义时空图）、自我改进（自举纪律）、零LLM白箱管线与可审计信任护栏。

NVIDIA Cosmos is an open platform of world models, datasets, and tools that enables developers to build Physical AI for robots, autonomous vehicles, smart infrastructure, and more.

Datasets use the HDF5 format for fast loading. Download the data from HuggingFace and decompress with:

Every clip was produced by the launchers in this repo, on the data bundled in examples/ — no external dataset, no cherry-picking across seeds. mp4.

Code World Model Official Repo

5 is an efficient causal world model built for real-time interaction. It continuously rolls out a visual world from its current state while text prompts and keyboard actions alter the world's semantics, motion, and future state as it unfolds.

5 FM → Stage1 TF-AnyFlow → Stage2 DMD via SGF, turning a bidirectional video model into a camera-controlled few-step autoregressive model.

A unified framework for training, fine-tuning, and evaluating World Action Models

H3-World is the first interactive world model built on MiniMax-H3.

IMPORTANT] > ## 🎉 LingBot-World-Infinity is now available! > > This repository is no longer actively maintained. > > 👉 Please use the latest repository: > > > All future updates, new models, and new features will be released in LingBot-World-Infinity.

\ Cosmos straddles Renderer and Simulator. That ambiguity is the point. Before asking whether something is a world model, ask what it predicts and what you can do with the why-prediction-is-learning.

Motus2: A Self-Evolving General World Model for Dexterous Manipulation

1 agent swarms take a brief - a sentence, a photograph, a clip - then research the place, model it, render it, and check it. js app that opens in a browser.

The in-context human video generation pipeline starts from task-sampled robot trajectories that retain executable actions.

Pixels are evidence of the physical world, not its ontology. A pixel-level observation records how the world appears at a particular moment and from a particular viewpoint, but does not directly specify what exists within it, how it is structured, or what governs its evolution.

Flex-π is a 6B-parameter world-action model for robot manipulation. It jointly predicts future RGB, 3D pointmaps, DINOv3 semantics, and actions in training, then deploys as a VLA, as a full world-action model, or as anything in between — all from a single checkpoint.

Video diffusion transformers address their tokens by position on the pixel-time grid: an address in the tensor, not in the world.

A high-performance serving framework for robotics models.

This project has not provided a description yet.

Upstream provenance for observation/ — the exact commit, what was modified, and what was removed — is in observation/PROVENANCE.md.

This repo contains the harness and six complete session records. The full post-broker record will be released shortly. The score by itself is not the interesting part: several systems built specifically for ARC score this high or higher.