# kzl/decision-transformer

Official codebase for Decision Transformer: Reinforcement Learning via Sequence Modeling.

Repository: https://github.com/kzl/decision-transformer
Canonical: https://ross.abutalabs.com/products/decision-transformer
Language: Python
License: MIT
License Family: permissive
Last push: 2024-04-29T01:30:45+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1918, "days_push": 857, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2830, forks 517 (observed 2026-08-28T04:07:24.303834+00:00)

## What it is
Official codebase for the Decision Transformer paper, which frames reinforcement learning as sequence modeling with transformers. It contains scripts to reproduce Atari and OpenAI Gym experiments from the paper.

## Use cases
- reproduce Decision Transformer paper experiments
- train a transformer for offline reinforcement learning
- apply sequence modeling to Atari environments
- run RL experiments on OpenAI Gym tasks
- study transformers as an alternative to traditional RL algorithms

## When to choose
- you want to reproduce or build on the Decision Transformer paper
- you need an offline RL baseline using sequence modeling
- you are researching transformer-based RL approaches

## When to avoid
- you need a production-ready RL framework
- you want online RL or modern maintained tooling
- you need environments beyond Atari and Gym without adaptation

## Facets
- artifact type: library
- maturity: maintenance
- function: reinforcement-learning, deep-learning, machine-learning
- domain: reinforcement-learning, deep-learning, machine-learning
- platform: python
- tags: transformer, sequence-modeling, research-code, atari, openai-gym, paper-reproduction, research, linux, macos

## Member repositories
- kzl/decision-transformer (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:24.303834+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 2026-08-29T18:48:39.881080+00:00, confidence not recorded.
  - readme: https://github.com/kzl/decision-transformer (fetched 2026-08-28T04:07:24.303834+00:00, sha ae33c97aa0a5)
- Data as of 2026-08-30T08:39:29.467469+00:00.