# rl-tools/rl-tools

The Fastest Deep Reinforcement Learning Library

Repository: https://github.com/rl-tools/rl-tools
Canonical: https://ross.abutalabs.com/products/rl-tools
Homepage: https://rl.tools
Language: C++
License: MIT
License Family: permissive
Topics: continuous-control, cpp, deep-learning, mujoco, reinforcement-learning, robotics, tinyml, tinyrl
Last push: 2026-07-04T19:06:26+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 90, release rhythm 29, longevity 73
- inputs: {"age_days": 1027, "days_push": 60, "days_rel": 314, "gap_med": 168.5, "n_releases_24m": 3}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1028, forks 60 (observed 2026-08-28T04:03:17.497957+00:00)

## What it is
RLtools is a pure C++ header-only, dependency-free deep reinforcement learning library supporting algorithms like SAC, TD3, and PPO. It compiles to WASM to run in browsers and targets a broad range of devices including embedded hardware.

## Use cases
- train deep RL agents for continuous control fast
- run reinforcement learning in the browser via wasm
- train RL policies on embedded or tiny devices
- simulate and train on MuJoCo environments
- learn drone or robot control policies in seconds on a laptop

## When to choose
- you need maximum training throughput for continuous-control RL
- you want a dependency-free C++ RL library deployable to embedded or browser targets

## When to avoid
- you need a high-level Python-first RL framework with a large algorithm ecosystem
- you work mainly with discrete/game environments rather than continuous control

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, reinforcement-learning, simulation
- domain: reinforcement-learning, machine-learning, robotics, deep-learning
- platform: cpp, wasm, browser, cross-platform
- tags: header-only, continuous-control, mujoco, tinyml, sac, td3, ppo, robotics, embedded, gpu

## Member repositories
- rl-tools/rl-tools (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:17.497957+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 2026-08-30T07:07:49.024554+00:00, confidence not recorded.
  - readme: https://github.com/rl-tools/rl-tools (fetched 2026-08-28T04:03:17.497957+00:00, sha 4d73f18cf5a8)
  - homepage: https://rl.tools (fetched 2026-08-29T13:07:47.628260+00:00, sha 4c7194210c2b)
  - site_page: https://docs.rl.tools (fetched 2026-08-29T13:07:47.630786+00:00, sha 44136fa355b3)
- Data as of 2026-08-30T08:39:29.467469+00:00.