A curated list of AI tools and resources for developers, see the AI Resources .

Gymnasium

An API standard for single-agent reinforcement learning environments, with popular reference environments and related utilities (formerly Gym).

Overview

Gymnasium provides an API standard for single-agent reinforcement learning environments, offering popular reference environments and related utilities as a modern successor to OpenAI Gym. The project serves as a stable baseline for training, evaluating, and researching RL algorithms.

Key Features

  • Standardized API: Unified environment interface for reproducible experiments and algorithm comparisons.
  • Reference environments: Includes many common RL environments and supporting utilities.
  • Community & ecosystem: Maintained by the Farama Foundation with active contributor engagement.

Use Cases

  • RL research: Quickly set up training and evaluation baselines for experiments.
  • Teaching & demos: Reproducible environments for classroom examples and algorithm instruction.
  • Simulation & benchmarking: Standardized platform for comparing training strategies and algorithm performance.

Technical Details

  • Stack: Python-first ecosystem compatible with mainstream RL toolchains and dependencies.
  • Extensibility: Environments and tools are easy to extend and adapt to new scenarios.
  • License: MIT license suitable for research and commercial use.

Comments

Gymnasium
Resource Info
🏋️ Training 🕹️ Simulator 🌱 Open Source