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.
A simple and highly efficient RTS-game-inspired environment for reinforcement learning (formerly Gym-MicroRTS)
| Date | Stars |
|---|---|
| 2026-07-31 | 289 |
| 2026-08-01 | 289 |
| 2026-08-02 | 289 |
| 2026-08-06 | 289 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
<p align="center">
<img src="https://raw.githubusercontent.com/Farama-Foundation/MicroRTS-Py/master/micrortspy-text.png" width="500px"/>
</p>
**Aug 11, 2025: This project has been deprecated due to a lack of wide spread community use, and is no longer planned to receive any additional updates or support.**
Formerly Gym-μRTS/Gym-MicroRTS
[<img src="https://img.shields.io/badge/discord-gym%20microrts-green?label=Discord&logo=discord&logoColor=ffffff&labelColor=7289DA&color=2c2f33">](https://discord.gg/DdJsrdry6F)
[<img src="https://github.com/vwxyzjn/gym-microrts/workflows/build/badge.svg">](https://github.com/Farama-Foundation/MicroRTS-Py/actions)
[<img src="https://badge.fury.io/py/gym-microrts.svg">](
https://pypi.org/project/gym-microrts/)
This repo contains the source code for the gym wrapper of μRTS authored by [Santiago Ontañón](https://github.com/santiontanon/microrts).
MicroRTS-Py will eventually be updated, maintained, and made compliant with the standards of the Farama Foundation (https://farama.org/project_standards). However, this is currently a lower priority than other projects we're working to maintain. If you'd like to contribute to development, you can join our discord server here- https://discord.gg/jfERDCSw.
## Get Started
Prerequisites:
* Python 3.8+
* [Poetry](https://python-poetry.org)
* Java 8.0+
* FFmpeg (for video recording utilities)
```bash
$ git clone --recursive https://github.com/Farama-Foundation/MicroRTS-Py.git && \
cd MicroRTS-Py
poetry install
# The `poetry install` command above creates a virtual environment for us, in which all the dependencies are installed.
# We can use `poetry shell` to create a new shell in which this environment is activated. Once we are done working with
# MicroRTS, we can leave it again using `exit`.
poetry shell
# By default, the torch wheel is built with CUDA 10.2. If you are using newer NVIDIA GPUs (e.g., 3060 TI), you may need to specifically install CUDA 11.3 wheels by overriding the torch dependency with pip:
# poetry run pip install "torch==1.12.1" --upgrade --extra-index-url https://download.pytorch.org/whl/cu113
python hello_world.py
```
If the `poetry install` command gets stuck on a Linux machine, [it may help to first run](https://github.com/python-poetry/poetry/issues/8623): `export PYTHON_KEYRING_BACKEND=keyring.backends.null.Keyring`.
To train an agent, run the following
```bash
cd experiments
python ppo_gridnet.py \
--total-timesteps 100000000 \
--capture-video \
--seed 1
```
[](https://asciinema.org/a/586754)
For running a partial observable example, tune the `partial_obs` argument.
```bash
cd experiments
python ppo_gridnet.py \
--partial-obs \
--capture-video \
--seed 1
```
## Technical Paper
Before diving into the code, we highly recommend reading the preprint of our paper: [Gym-μRTS: Toward Affordable Deep Reinforcement Learning Research in Real-time Strategy Games](https://arxiv.org/abs/2105.13807).
### Depreciation notes
1. Note that the experiments in the technical paper above are done with [`gym_microrts==0.3.2`](https://github.com/vwxyzjn/gym-microrts/tree/v0.3.2). As we move forward beyond `v0.4.x`, we are planning to deprecate UAS despite its better performance in the paper. This is because UAS has a more complex implementation and makes it really difficult to incorporate selfplay or imitation learning in the future.
2. [v0.6.1](https://github.com/Farama-Foundation/MicroRTS-Py/releases/tag/v0.6.1) is the last version in which wall/terrain observations were not present in state tensors. As of December 2023, every state observation has an extra channel encoding the presence of walls, and models trained before this will therefore no longer be compatible with code in the `master` branch. Such models should use the code from `v0.6.1` instead.
## EExcerpt of 14,368 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:e86c2c0398e1519a, llm:description: 'A simple and highly efficient RTS-game-inspired environment for reinforcement learning (formerly Gym-MicroRTS)'
matched fp:e86c2c0398e1519a, llm:description: 'A simple and highly efficient RTS-game-inspired environment for reinforcement learning (formerly Gym-MicroRTS)'