Skip to content
RobothecaBeta
Log in

Stable-Baselines3

  • Software
  • Humanoids and robot learning
  • Library

Stable-Baselines3 is a set of reliable PyTorch implementations of reinforcement learning algorithms, meant as baselines for research and for building new projects.

Made by DLR-RM.

Source on GitHub: github.com/DLR-RM/stable-baselines3 (opens in a new tab). Facts read at commit b2878f4 (opens in a new tab).

Licenses

Code
MIT (opens in a new tab)

Read by Robotheca from LICENSE (opens in a new tab) (MIT) at commit b2878f4 (opens in a new tab), on 16 Oct 2026.

Commercial use
Yes

What it does

  • Provides implementations of many reinforcement learning algorithms, including A2C, DDPG, DQN, PPO, SAC, TD3 and HER, with a common interface.
  • Supports custom environments and custom policies, plus Dict observation spaces and TensorBoard logging.
  • Offers a sklearn-like syntax for training and running agents, for example PPO on CartPole in a few lines.
  • Experimental and newer algorithms live in the separate SB3-Contrib and SBX (Jax) repositories.
  • Includes the RL Baselines3 Zoo, a training framework with tuned hyperparameters and trained agents.
  • Assumes some prior knowledge of reinforcement learning; the project does not offer technical support.

Written by Robotheca from the project's README (opens in a new tab) at commit b2878f4.

Getting started

Difficulty
Intermediate
Hardware it needs, as the project states it
Requires Python 3.10+ and PyTorch >= 2.8; no GPU, memory or board requirements are stated.

Maintenance

Status
Active: a commit in the last 90 days
Newest commit
13 Oct 2026 on master
Latest release
v2.9.0: Updated dependencies (pandas is now optional, gymnasium 1.3.0 support, torch>=2.8) (15 Jun 2026)

Checked on GitHub on 16 Oct 2026.

Links

Robots on Robotheca that use it

No robot on Robotheca is linked to it yet. Links come from a robot's robot.yaml, from its dependency files read as text, or from Robotheca's editors.

Robotheca lists software as a reference: nothing is hosted or run here, and every fact links to where it was read.