Metadata-Version: 2.4
Name: embodichain
Version: 0.3.2
Summary: An end-to-end, GPU-accelerated, and modular platform for building generalized Embodied Intelligence.
Home-page: https://github.com/DexForce/EmbodiChain
Author: EmbodiChain Developers
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: dexsim_engine==0.5.1rc1
Requires-Dist: shapely>=2.0
Requires-Dist: setuptools>=78.1.1
Requires-Dist: gymnasium>=0.29.1
Requires-Dist: huggingface_hub<1.0,>=0.35.3
Requires-Dist: langchain
Requires-Dist: langchain-openai
Requires-Dist: scipy
Requires-Dist: pin
Requires-Dist: pin-pink
Requires-Dist: casadi
Requires-Dist: qpsolvers[osqp]==4.8.1
Requires-Dist: pytorch_kinematics==0.10.0
Requires-Dist: polars==1.31.0
Requires-Dist: PyYAML>=6.0
Requires-Dist: wandb>=0.21.0
Requires-Dist: tensorboard>=2.20.0
Requires-Dist: ortools
Requires-Dist: prettytable
Requires-Dist: psutil>=5.9
Requires-Dist: black==26.10.1
Requires-Dist: fvcore
Requires-Dist: h5py
Requires-Dist: tensordict
Requires-Dist: viser==1.0.21
Requires-Dist: lerobot<0.5,>=0.4.4
Requires-Dist: torchcodec<0.6,>=0.5
Provides-Extra: policy-deploy
Requires-Dist: onnxruntime-gpu<1.27,>=1.20; extra == "policy-deploy"
Provides-Extra: gensim
Requires-Dist: bpy; extra == "gensim"
Requires-Dist: gradio<6.18,>=6.17.3; extra == "gensim"
Requires-Dist: pyrender==0.1.45; extra == "gensim"
Requires-Dist: requests; extra == "gensim"
Requires-Dist: Pillow; extra == "gensim"
Requires-Dist: matplotlib; extra == "gensim"
Dynamic: home-page
Dynamic: license-file

# EmbodiChain

![teaser](assets/imgs/teaser.jpg)

[![License](https://img.shields.io/github/license/DexForce/EmbodiChain?style=for-the-badge)](LICENSE)
[![Website](https://img.shields.io/badge/website-dexforce.com-yellow?style=for-the-badge&logo=google-chrome&logoColor=white)](https://dexforce.com/embodichain/index.html#/)
[![GitHub Pages](https://img.shields.io/badge/GitHub%20Pages-docs-blue?style=for-the-badge&logo=github&logoColor=white)](https://dexforce.github.io/EmbodiChain/main/index.html)
[![Python](https://img.shields.io/badge/python-3.10%20|%203.11%20|%203.12-blue?style=for-the-badge&logo=python&logoColor=white)](https://docs.python.org/3/whatsnew/3.12.html)
[![Version](https://img.shields.io/github/v/release/DexForce/EmbodiChain?style=for-the-badge&label=version)](https://github.com/DexForce/EmbodiChain/releases)
---

EmbodiChain is an end-to-end, GPU-accelerated framework for Embodied AI. It streamlines research and development by unifying high-performance simulation, automated generative data pipelines, modular model architectures, and efficient training workflows. This integration enables rapid experimentation, seamless deployment of intelligent agents, and effective Sim2Real transfer for real-world robotic systems.

Official task and robot modules live under [extensions/](extensions/README.md)
and ship with the main distribution. Their Python imports remain
`embodichain_tasks` and `embodichain_robots`.

> [!NOTE]
> EmbodiChain is in Alpha and under active development:
> * More features will be continually added in the coming months. You can find more details in the [roadmap](https://dexforce.github.io/EmbodiChain/main/resources/roadmap.html).
> * Since this is an early release, we welcome feedback (bug reports, feature requests, etc.) via GitHub Issues.


## Key Features

- 🚀 **High-Fidelity GPU Simulation**: Realistic physics for rigid & deformable objects, advanced ray-traced sensors, all GPU-accelerated for high-throughput batch simulation.
- 🤖 **Unified Robot Learning Environment**: Standardized interfaces for Imitation Learning, Reinforcement Learning, and more.
- 📊 **Scalable Data Pipeline**: Automated data collection, efficient processing, and large-scale generation for model training.
- ⚡ **Efficient Training & Evaluation**: Online data streaming, parallel environment rollouts, and modern training paradigms.
- 🧩 **Modular & Extensible**: Easily integrate new robots, environments, and learning algorithms.

The figure below illustrates the overall architecture of EmbodiChain:

<p align="center">
  <img src="assets/imgs/frameworks.jpg" alt="architecture" width="90%"/>
</p>

Explore EmbodiChain's module relationships with the [interactive architecture visualization](https://dexforce.github.io/EmbodiChain/main/overview/architecture/index.html#).


## Getting Started

To get started with EmbodiChain, follow these steps:

- [Installation Guide](https://dexforce.github.io/EmbodiChain/main/quick_start/install.html)
- [Quick Start Tutorial](https://dexforce.github.io/EmbodiChain/main/tutorial/index.html)
- [API Reference](https://dexforce.github.io/EmbodiChain/main/api_reference/index.html)

### Task Environments

Official task environments are included in the main `embodichain` installation.
Use `embodichain list-task` to discover tasks and `embodichain run-task` to launch
one with its configuration. See [Supported Tasks](https://dexforce.github.io/EmbodiChain/main/resources/task/index.html)
for available tasks, runnable configurations, and task authoring guides.

For external training, see the optional
[RLinf integration](https://dexforce.github.io/EmbodiChain/main/guides/rlinf.html)
and the [official RLinf EmbodiChain recipe](https://rlinf.readthedocs.io/en/latest/rst_source/examples/embodied/embodichain.html).

## Contribution Guide

We welcome contributions! Please see the [CONTRIBUTING.md](https://github.com/DexForce/EmbodiChain/blob/main/CONTRIBUTING.md) file in this repository for guidelines on how to get started.

## Publications

See [Academic Publications](https://dexforce.github.io/EmbodiChain/main/resources/publications/README.html) for a complete list of academic papers related to EmbodiChain.

## Citation

If you find EmbodiChain helpful for your research, please consider citing our work:

```bibtex
@misc{EmbodiChain,
  author = {EmbodiChain Developers},
  title = {EmbodiChain: An end-to-end, GPU-accelerated, and modular platform for building generalized Embodied Intelligence},
  month = {November},
  year = {2025},
  url = {https://github.com/DexForce/EmbodiChain}
}
```

```bibtex
@misc{GS-World,
   author = {Guiliang Liu and Yueci Deng and Zhen Liu and Kui Jia},
   title = {GS-World: An Efficient, Engine-driven Learning Paradigm for Pursuing Embodied Intelligence using World
      Models of Generative Simulation},
   month = {October},
   year = {2025},
   journal = {TechRxiv}
}
```
