# kengz/SLM-Lab

Modular Deep Reinforcement Learning framework in PyTorch. Companion library of the book "Foundations of Deep Reinforcement Learning".

Repository: https://github.com/kengz/SLM-Lab
Canonical: https://ross.abutalabs.com/products/slm-lab
Homepage: https://slm-lab.gitbook.io/slm-lab/
Language: Python
License: MIT
License Family: permissive
Topics: pytorch, reinforcement-learning, deep-reinforcement-learning, benchmark, policy-gradient, dqn, ppo, sac, a2c, a3c
Last push: 2026-08-20T22:18:32+00:00

## Health v2 (maintenance only)
Score: 95/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 89, longevity 100
- inputs: {"age_days": 3257, "days_push": 13, "days_rel": 74, "gap_med": 8, "n_releases_24m": 6}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1362, forks 290 (observed 2026-08-28T04:04:30.305035+00:00)

## What it is
SLM Lab is a modular deep reinforcement learning framework built in PyTorch, offering ready-to-use algorithms like PPO, SAC, DQN, and A2C validated on 70+ environments. It serves as the companion library to the book 'Foundations of Deep Reinforcement Learning' and includes benchmarking, configuration via JSON specs, and GPU/async training support.

## Use cases
- train a PPO agent on CartPole
- run reinforcement learning benchmarks on Atari environments
- learn deep RL by following a book with runnable code
- benchmark SAC on continuous control tasks like HalfCheetah
- run async RL training with Hogwild! on GPU
- compare policy gradient algorithms across environments

## When to choose
- you want a modular PyTorch RL framework with validated algorithms
- you are studying deep RL and want code matching a textbook
- you need reproducible RL benchmarks with public data

## When to avoid
- you need a production RL deployment platform rather than a research framework
- you work outside Python/PyTorch ecosystems

## Facets
- artifact type: framework
- maturity: active
- function: machine-learning, reinforcement-learning, benchmarking
- domain: reinforcement-learning, machine-learning, deep-learning, gaming-tools, tutorials
- platform: python, cross-platform
- tags: pytorch, deep-reinforcement-learning, ppo, sac, dqn, a2c, policy-gradient, gymnasium, atari, benchmarking, education, companion-book, gpu

## Member repositories
- kengz/SLM-Lab (main) score 95

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:30.305035+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 2026-08-30T04:41:30.890668+00:00, confidence not recorded.
  - readme: https://github.com/kengz/SLM-Lab (fetched 2026-08-28T04:04:30.305035+00:00, sha 81082d4f1e47)
  - homepage: https://slm-lab.gitbook.io/slm-lab/ (fetched 2026-08-29T11:59:09.457322+00:00, sha 33f789349813)
- Data as of 2026-08-30T08:39:29.467469+00:00.