mees/calvin

CALVIN - A benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks

48
/ 100
Emerging

This project provides a standardized way to train and evaluate AI models for complex robot manipulation. It takes human language instructions, along with various sensor inputs like camera images or tactile data, and helps create robotic agents that can perform long sequences of actions, like opening a drawer and placing an object inside. It's designed for robotics researchers and engineers who are developing AI for practical, language-guided robot tasks.

854 stars. No commits in the last 6 months.

Use this if you are a robotics researcher working on developing AI agents that can understand and execute multi-step robot manipulation tasks based on natural language commands.

Not ideal if you need a plug-and-play solution for immediate robot deployment, as this is a benchmark for developing and testing new AI policy learning methods.

robotics research robot manipulation language understanding AI policy learning robot control
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 20 / 25

How are scores calculated?

Stars

854

Forks

111

Language

Python

License

MIT

Last pushed

Sep 08, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mees/calvin"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.