taco-group/OpenEMMA
OpenEMMA is an open-source end-to-end autonomous driving framework that uses pretrained Vision Language Models to predict ego waypoints from camera and text inputs.
Not currently ranked — collecting fresh signals.
star history
OpenEMMA is an open-source reproduction of Waymo’s EMMA model for autonomous driving. It leverages pretrained VLMs such as GPT-4 and LLaVA to integrate front-view camera images and text inputs, enabling real-time prediction of future ego waypoints along with decision rationales. The project provides a PyPI-installable package and an end-to-end pipeline for motion planning in autonomous vehicles.
Frequently asked
- What is taco-group/OpenEMMA?
- OpenEMMA is an open-source end-to-end autonomous driving framework that uses pretrained Vision Language Models to predict ego waypoints from camera and text inputs.
- Is OpenEMMA open source?
- Yes — taco-group/OpenEMMA is open source, released under the Apache-2.0 license.
- What language is OpenEMMA written in?
- taco-group/OpenEMMA is primarily written in Python.
- How popular is OpenEMMA?
- taco-group/OpenEMMA has 946 stars on GitHub.
- Where can I find OpenEMMA?
- taco-group/OpenEMMA is on GitHub at https://github.com/taco-group/OpenEMMA.