OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Reaver

Reaver is an MIT-licensed Python framework for developing modular deep reinforcement-learning agents, focused on StarCraft II tasks and also supporting Gym, Atari and MuJoCo environments.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Reaver — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Reaver when exploring modular deep reinforcement-learning experiments for StarCraft II or its listed alternative environments, and when you can manage dependencies and project maintenance yourself.

Who it is for

  • Hobbyist programmers experimenting with reinforcement-learning agents and hyperparameters.
  • Researchers who want to swap agent, model and environment components in experiments.

Documented uses

  • Develop and train agents for StarCraft II scenarios.
  • Run A2C or other documented agent experiments using configurable training options.
  • Adapt experiments to Gym, Atari or MuJoCo by installing the relevant environment packages.
  • Modify or compare modular agent, model and environment components.
Before you choose

Requirements & limits

The public repository specifies an MIT license. Its README says the project is no longer maintained and that the author cannot provide further development or support.

  • StarCraft II use requires the README’s listed PySC2, StarCraft II, gin-config, TensorFlow and TensorFlow Probability versions.
  • Gym, Atari and MuJoCo require additional packages; Roboschool is listed as an alternative for MuJoCo.
  • The README recommends Linux for performance and stability, while providing separate Windows setup instructions.
  • The project is explicitly no longer maintained, which matters when planning dependency updates or seeking support.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install Reaver for StarCraft II experiments

Set up Reaver using the README’s package installation route and identify its required StarCraft II dependencies.

Before you start

  • Python and pip, as implied by the documented pip installation command.
  • PySC2 >= 3.0.0 and StarCraft II >= 4.1.2.
  • gin-config >= 0.3.0, TensorFlow >= 2.0.0, and TensorFlow Probability >= 0.9.

Steps

  1. Install Reaver with the command below.

    Documented command
    pip install reaver

  2. Install the listed StarCraft II, PySC2, configuration and TensorFlow dependencies at the specified minimum versions.

  3. For Gym, Atari or MuJoCo environments, install the relevant optional packages listed by the README.

Expected outcome

Reaver and the listed dependencies are installed as described; the README provides a separate quick-start example for training.

Guide limitations

  • The README says the project is no longer maintained and the author cannot provide further development or support.
  • The guide covers installation only; it does not verify that the commands or dependencies work in a particular environment.
  • Gym, Atari and MuJoCo use additional packages; Roboschool is listed as a MuJoCo alternative.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
inoryy/reaver · Official homepage title · retained in its source language
How we review sourcesSuggest a correction