Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
JAX-accelerated Meta-Reinforcement Learning Environments Inspired by XLand and MiniGrid 🏎️
| Date | Stars |
|---|---|
| 2026-07-31 | 343 |
| 2026-08-04 | 343 |
| 2026-08-05 | 343 |
| 2026-08-06 | 343 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
<h1 align="center">XLand-MiniGrid</h1>
<p align="center">
<a href="https://pypi.python.org/pypi/xminigrid">
<img src="https://img.shields.io/pypi/pyversions/xminigrid.svg"/>
</a>
<a href="https://badge.fury.io/py/xminigrid">
<img src="https://badge.fury.io/py/xminigrid.svg"/>
</a>
<a href="https://github.com/astral-sh/ruff">
<img src="https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/charliermarsh/ruff/main/assets/badge/v2.json"/>
</a>
<a href="https://arxiv.org/abs/2312.12044">
<img src="https://img.shields.io/badge/arXiv-2210.07105-b31b1b.svg"/>
</a>
<a href="https://twitter.com/vladkurenkov/status/1731709425524543550">
<img src="https://badgen.net/badge/icon/twitter?icon=twitter&label"/>
</a>
<a target="_blank" href="https://colab.research.google.com/github/corl-team/xland-minigrid/blob/main/examples/walkthrough.ipynb">
<img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/>
</a>
</p>
[//]: # ( <a href="https://badge.fury.io/py/xminigrid">)
[//]: # ( <img src="https://img.shields.io/pypi/dm/xminigrid?color=yellow&label=Downloads"/>)
[//]: # ( </a>)
[//]: # ( <a href="https://github.com/corl-team/xland-minigrid/main/LICENSE">)
[//]: # ( <img src="https://img.shields.io/badge/license-Apache_2.0-blue"/>)
[//]: # ( </a>)
# Meta-Reinforcement Learning in JAX
<!-- > 🥳 We recently released [**XLand-100B**](https://github.com/dunno-lab/xland-minigrid-datasets), a large multi-task dataset for offline meta and in-context RL research, based on XLand-MiniGrid.
It is currently the largest dataset for in-context RL, containing full learning histories for **30k unique tasks, 100B transitions, and 2.5B episodes**. Check it out! -->
**XLand-MiniGrid** is a suite of tools, grid-world environments and benchmarks for meta-reinforcement learning research inspired by
the diversity and depth of [XLand](https://deepmind.google/discover/blog/generally-capable-agents-emerge-from-open-ended-play/)
and the simplicity and minimalism of [MiniGrid](https://github.com/Farama-Foundation/MiniGrid). Despite the similarities,
XLand-MiniGrid is written in JAX from scratch and designed to be highly scalable, democratizing large-scale experimentation
with limited resources. Ever wanted to reproduce a [DeepMind AdA](https://sites.google.com/view/adaptive-agent/) agent? Now you can and not in months, but days!
### Features
- 🔮 System of rules and goals that can be combined in arbitrary ways to produce
diverse task distributions
- 🔧 Simple to extend and modify, comes with example environments ported from the original
[MiniGrid](https://github.com/Farama-Foundation/MiniGrid)
- 🪄 Fully compatible with all JAX transformations, can run on CPU, GPU and TPU
- 📈 Easily scales to $2^{16}$ parallel environments and millions of steps per second on a single GPU
- 🔥 Multi-GPU PPO baselines in the [PureJaxRL](https://github.com/luchris429/purejaxrl) style, which can achieve **1 trillion** environment steps under two days
How cool is that? For more details, take a look at the [technical paper](https://arxiv.org/abs/2312.12044) or
[examples](examples), which will walk you through the basics and training your own adaptive agents in minutes!
[//]: # ()
## Installation 🎁
The latest release of XLand-MiniGrid can be installed directly from PyPI:
```commandline
uv add xminigrid (or pip install xminigrid)
# or, from github directly
uv pip install "xminigrid @ git+https://github.com/corl-team/xland-minigrid.git"
```
Alternatively, if you want to install the latest development version from the GitHub and run provided algorithms or scripts,
install the source as follows:
```commandline
git clone [email protected]:corl-team/xland-minigrid.git
cd xland-minigrid
# additional dependencies for dev and baselines
uv syExcerpt of 17,278 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:79854685b9fe0d34, llm:Repository topics and description: 'jax, meta-reinforcement-learning, minigrid, reinforcement-learning, xland' and description: 'JAX-accelerated Meta-Reinforcement Learning Environments Inspired by XLand and MiniGrid'.
matched fp:79854685b9fe0d34, llm:Repository topics and description: 'jax, meta-reinforcement-learning, minigrid, reinforcement-learning, xland' and description: 'JAX-accelerated Meta-Reinforcement Learning Environments Inspired by XLand and MiniGrid'.
matched fp:79854685b9fe0d34, llm:Repository topics and description: 'jax, meta-reinforcement-learning, minigrid, reinforcement-learning, xland' and description: 'JAX-accelerated Meta-Reinforcement Learning Environments Inspired by XLand and MiniGrid'.