X Transformers Rl: Train Agents in Simulators

X Transformers Rl

X Transformers Rl is a work-in-progress Python library for reinforcement-learning experiments using x-transformers. Its Learner interface accepts environment dimensions and world-model settings, then trains against a supplied simulator.

Independent OpenViST project profile, based on linked sources.

Understand the project

Practical overview

When it fits

Use it when building or exploring a Python reinforcement-learning experiment around a simulator and a transformer-based world-model configuration.

Who it is for

Python developers prototyping reinforcement-learning agents in simulators.
Researchers exploring transformer-based world-model configurations.
Developers adapting the documented Lunar Lander example.

Documented uses

Configure a learner with state and action dimensions, reward range and a timestep limit.
Set world-model parameters such as attention head dimension, number of heads and depth.
Pass a simulator to the learner for a training run.

Before you choose

Requirements & limits

The project is distributed as a Python package installable with pip. The public repository specifies the MIT license.

The README labels the project work in progress.
The documented learner interface expects a simulator and requires the developer to supply environment dimensions and related parameters.
The repository provides a Lunar Lander example in addition to a generic simulator pattern.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up the documented Lunar Lander example

Install the example requirements and start its training script.

Before you start

Python environment able to install the project's listed requirements.

Steps

From the project repository, install dependencies with the command below.
Documented command
```
pip install -r requirements.txt
```
Run the command below.
Documented command
```
python train_lander.py
```

Expected outcome

The documented Lunar Lander training script is started.

Guide limitations

The README does not specify expected training duration, results or additional environment requirements.

Documentation 1 ↗

Evidence and transparency

Sources & review

Official repository ↗

Source record details

Source availability: Active
Sources checked: 15 October 2026
Original website title: lucidrains/x-transformers-rl · Official homepage title · retained in its source language

How we review sources Suggest a correction