# DLR-RM/stable-baselines3

PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.

Repository: https://github.com/DLR-RM/stable-baselines3
Canonical: https://ross.abutalabs.com/products/stable-baselines3
Homepage: https://stable-baselines3.readthedocs.io
Language: Python
License: MIT
License Family: permissive
Topics: reinforcement-learning, reinforcement-learning-algorithms, machine-learning, gym, openai, baselines, toolbox, stable-baselines, python, pytorch, robotics, sde, gsde, sb3
Last push: 2026-08-17T16:40:34+00:00

## Health v2 (maintenance only)
Score: 91/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 98, release rhythm 77, longevity 100
- inputs: {"age_days": 2311, "days_push": 16, "days_rel": 79, "gap_med": 75, "n_releases_24m": 8}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 13728, forks 2171 (observed 2026-08-28T04:11:04.679160+00:00)

## What it is
Stable Baselines3 is a Python library providing reliable, well-tested PyTorch implementations of reinforcement learning algorithms with a unified API. It serves as a standard baseline toolbox for RL research and applications, supporting custom environments, policies, and TensorBoard logging.

## Use cases
- train reinforcement learning agents in python
- compare my RL algorithm against standard baselines
- implement PPO or SAC without writing it from scratch
- train agents on custom gymnasium environments
- learn reinforcement learning with a simple high-level library
- benchmark RL algorithm performance

## When to choose
- you want trusted, tested PyTorch RL algorithm implementations with a common interface
- you need reproducible baselines for RL research
- you want beginner-friendly access to state-of-the-art RL methods

## When to avoid
- you need the very latest experimental RL algorithms (see SB3 Contrib or SBX)
- you have no prior reinforcement learning knowledge and want a fully guided tutorial
- you need JAX-based training rather than PyTorch

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, reinforcement-learning, benchmarking
- domain: reinforcement-learning, machine-learning, robotics
- platform: python, cross-platform
- tags: pytorch, gymnasium, rl-algorithms, baselines, openai-gym, training-agents

## Member repositories
- DLR-RM/stable-baselines3 (main) score 91

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:04.679160+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:13:05.095861+00:00, confidence not recorded.
  - readme: https://github.com/DLR-RM/stable-baselines3 (fetched 2026-08-28T04:11:04.679160+00:00, sha 7a712837e38c)
  - homepage: https://stable-baselines3.readthedocs.io (fetched 2026-08-29T08:07:26.106497+00:00, sha 8e4a84078a1e)
  - registry_pypi: https://pypi.org/pypi/stable-baselines3/json (fetched 2026-08-29T08:07:26.116095+00:00, sha 840c796475dd)
- Data as of 2026-08-30T08:39:29.467469+00:00.