# openai/spinningup

An educational resource to help anyone learn deep reinforcement learning.

Repository: https://github.com/openai/spinningup
Canonical: https://ross.abutalabs.com/products/spinningup
Homepage: https://spinningup.openai.com/
Language: Python
License: MIT
License Family: permissive
Last push: 2024-08-05T03:30:30+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2856, "days_push": 758, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 11915, forks 2464 (observed 2026-08-28T04:10:50.488176+00:00)

## What it is
Spinning Up in Deep RL is an OpenAI-produced educational resource combining tutorials, curated papers, and a Python code repo with short standalone implementations of key deep reinforcement learning algorithms (VPG, TRPO, PPO, DDPG, TD3, SAC). It is aimed at helping people build the background to become deep RL researchers.

## Use cases
- learn deep reinforcement learning from scratch
- understand how PPO and policy gradient algorithms work
- read clean minimal implementations of SAC and TD3
- find a curated list of important deep RL papers
- prepare for a career in RL research
- run RL experiments on Gym and MuJoCo environments
- practice RL implementation exercises

## When to choose
- you want a structured educational path into deep RL with theory and code
- you need readable reference implementations of core on-policy and off-policy algorithms
- you want algorithm documentation with derivations, pseudocode, and key papers

## When to avoid
- you need a production-grade or feature-rich RL training framework
- you need Windows support or the latest maintained algorithm implementations
- you need distributed or highly scalable RL training infrastructure

## Facets
- artifact type: learning-resource
- maturity: maintenance
- function: machine-learning, reinforcement-learning, developer-tools, documentation
- domain: reinforcement-learning, machine-learning, deep-learning, education, tutorials
- platform: python, cli
- tags: deep-rl, policy-gradient, openai-gym, mujoco, pytorch, educational, research-education, algorithm-implementations, linux, macos

## Member repositories
- openai/spinningup (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:50.488176+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 2026-08-29T17:15:12.728928+00:00, confidence not recorded.
  - readme: https://github.com/openai/spinningup (fetched 2026-08-28T04:10:50.488176+00:00, sha 4bc0a351d542)
  - homepage: https://spinningup.openai.com/ (fetched 2026-08-29T08:13:05.848003+00:00, sha 6a9484ed06b7)
  - site_page: https://spinningup.openai.com/en/latest/user/installation.html (fetched 2026-08-29T08:13:05.856852+00:00, sha a482aee64fb5)
  - site_page: https://spinningup.openai.com/en/latest/etc/author.html (fetched 2026-08-29T08:13:05.858825+00:00, sha 96833afc64b7)
  - site_page: https://spinningup.openai.com/en/latest/algorithms/vpg.html (fetched 2026-08-29T08:13:05.860598+00:00, sha c67dfd6c6bbe)
  - site_page: https://spinningup.openai.com/en/latest/algorithms/trpo.html (fetched 2026-08-29T08:13:05.862508+00:00, sha 57fec737a2bb)
  - site_page: https://spinningup.openai.com/en/latest/algorithms/ppo.html (fetched 2026-08-29T08:13:05.864470+00:00, sha 9028d78a3444)
  - site_page: https://spinningup.openai.com/en/latest/algorithms/ddpg.html (fetched 2026-08-29T08:13:05.866447+00:00, sha 39840f81d641)
  - site_page: https://spinningup.openai.com/en/latest/algorithms/td3.html (fetched 2026-08-29T08:13:05.868389+00:00, sha 72b3edb4ffe0)
  - site_page: https://spinningup.openai.com/en/latest/algorithms/sac.html (fetched 2026-08-29T08:13:05.870345+00:00, sha ab4a2cb54bc6)
- Data as of 2026-08-30T08:39:29.467469+00:00.