Understand the project
Practical overview
When it fits
Use Rt X when you want to explore or prototype with the listed RT-1-X or RT-2-X architectures in Python/PyTorch and can account for their documented implementation differences.
Who it is for
- Developers exploring PyTorch implementations of robotic-learning architectures.
- Researchers prototyping multimodal workflows using visual inputs and language.
Documented uses
- Inspect and prototype around RT-1-X using video and text instructions.
- Explore RT-2-X with image and text inputs and inspect its text-token output.
- Extract image features with the included EfficientNetFilm component.
Before you choose
Requirements & limits
The public GitHub repository is open source under the MIT license.
- RT-1-X examples use video and text instructions; RT-2-X examples use images and text.
- RT-2-X returns text tokens rather than a native seven-dimensional action vector.
- The RT-1-X backbone differs from the paper implementation: the repository describes a MaxViT substitution for FiLM-EfficientNet.
- The README says EfficientNet integration with RT-1-X is still in progress.
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
Install the package and find usage options
Install the Rt X package and consult the repository’s usage help.
Before you start
- Python environment with pip
Steps
Run the command below.
pip install rtx-torchRun the command below to view the usage options.
python run.py --help
Expected outcome
The package installation command has been run and the project’s usage help is available to consult.
Evidence and transparency
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 8 October 2026
- Original website title
- kyegomez/RT-X · Official homepage title · retained in its source language