Minerl

Visit Tool

MineRL is an AI Agents & Automation tool that provides a Python package for sample-efficient reinforcement learning research within Minecraft. It offers Gym environments and data access for training AI agents.

Claim this tool

No Views Yet

At a glance

Pricing

Open Source

Free tier

Yes

API

Yes

Skill level

Technical

About

What is minerl?

MineRL is a Python package designed for sample-efficient reinforcement learning research, primarily within the Minecraft environment. It provides easy-to-use Gym environments and data access, making it suitable for training AI agents. The package has evolved through several versions, with v1.0 supporting OpenAI VPT models and the MineRL BASALT 2022 competition, featuring a new Minecraft version (1.12 -> 1.16.5), larger default resolution (64x64 -> 640x360), and a near-human action-space focused on GUI and mouse control. It requires Java JDK 8 for installation and can be integrated into projects much like any standard Gym environment for developing and testing AI models.

Best used for

Ideal for researchers and developers who need to train AI agents in a complex, interactive environment like Minecraft, develop sample-efficient reinforcement learning algorithms, and participate in AI competitions. Especially valuable for those working with OpenAI VPT models or the MineRL BASALT competitions.

Common actions

train AI agents

develop reinforcement learning

access Minecraft data

github copilotface swappingworkflowslow-code/no-codecollaborationopen-sourceautomated workflowdeepfake"AI Agents"

Capabilities

Key features

Gym environments
Minecraft integration
Data access
Near-human action space
Pixel-only observation

Target Audience

professor

Integrations

Not yet documented

Pricing & Plans

Open Source

Free

FAQs

What are the key differences between MineRL versions?

MineRL has three main versions. v1.0 supports Minecraft 1.16.5 with larger resolution and a near-human action space. v0.4 was used for 2021 competitions and supports the original MineRL-v0 dataset. v0.3 was for earlier competitions (2019-2020) and also supports the original MineRL-v0 dataset.

What are the system requirements for installing MineRL?

To install MineRL, you need Java JDK 8. Mac users may require additional steps for proper setup. Once Java is installed, you can install the Python package using pip with the command `pip install git+https://github.com/minerllabs/minerl`.

Where can I find the MineRL-v0 dataset?

The original data mirrors for MineRL-v0 are currently down. However, copies of the primary datasets have been uploaded to Zenodo. You can find them at https://zenodo.org/records/12659939, which provides access to the necessary data for older MineRL versions.

Trending

Subcategories trending in AI Agents & Automation

Chatbots & Conversational AI General-Purpose Agents Workflow Agents Personal Assistants RAG & Document AI Voice Agents

Trending

Also listed in

This tool also appears in

Research & Education › Academic Research Coding & Development › Game Development

Explore

Browse AI tools by category

Content & Design Productivity & Business Coding & Development AI Agents & Automation Research & Education Wellness & Lifestyle Career Development Marketing & Growth Data & Analytics Customer Support & CX Finance E-commerce