pytorch/rl
A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.
This tool helps researchers and practitioners train Reinforcement Learning (RL) agents for various tasks without needing to write extensive code. You input specifications for an environment and desired agent behaviors, and it outputs a trained agent that can perform actions in that environment. It's designed for machine learning engineers, AI researchers, and data scientists looking to implement advanced decision-making systems.
3,335 stars. Actively maintained with 36 commits in the last 30 days.
Use this if you want to quickly train state-of-the-art Reinforcement Learning agents or fine-tune Large Language Models without deep coding knowledge, leveraging pre-built components and a command-line interface.
Not ideal if you need to build custom, low-level RL algorithms from scratch or integrate with non-PyTorch machine learning frameworks.
Stars
3,335
Forks
442
Language
Python
License
MIT
Category
Last pushed
Mar 13, 2026
Commits (30d)
36
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/pytorch/rl"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
DLR-RM/stable-baselines3
PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.
google-deepmind/dm_control
Google DeepMind's software stack for physics-based simulation and Reinforcement Learning...
Denys88/rl_games
RL implementations
yandexdataschool/Practical_RL
A course in reinforcement learning in the wild
Stable-Baselines-Team/stable-baselines3-contrib
Contrib package for Stable-Baselines3 - Experimental reinforcement learning (RL) code