Download AI Player — Minecraft Mods — MetaMods
AI Player

AI Player

Active

Downloads

9

Last update

1 month ago

Versions

1.20.1 — 1.21.1
Client
Fabric
Adventure
Social
Technological

AI Player

About the Modification

Have you ever felt lonely while playing Minecraft? This mod aims to solve this problem by adding an intelligent companion to the game that you can interact and communicate with.

The main advantage is that you can use local language models that work completely offline, depending on your computer's capabilities.

It's important to note that this is not a commercial AI product, but a solution to a problem many Minecraft players face.


What This Mod Does

The mod adds a second player-bot to the game that can be created through Carpet mod. You can communicate with this bot using large language models via Ollama completely offline. Reinforcement learning is also integrated, allowing the bot to interact with the environment and perform reflex actions.

Goal-based learning is planned for future updates.

Installation Requirements

For the mod to work, you need:

  • Latest version of Fabric API
  • Carpet mod version specified in the dependencies section
  • Java 21
  • Ollama for local models

Demonstration Videos

Overview of the mod in action:

Additional video demonstrating capabilities:

https://streamable.com/imko3x?src=player-page-share


Installation and Setup Guide

Step 1: Install Java 21

The project requires Java 21 to work:

  • Download Java 21 from the official website
  • On Windows: install the MSI package
  • On Linux: use the package manager, for example: sudo apt install openjdk-21-jdk

Step 2: Install Ollama

Go to https://ollama.com/ and download the version for your operating system. After installation, run Ollama - the server will run in the system tray.

Step 3: Configure Ollama Models

Open the command line and execute:

ollama pull nomic-embed-text   # Embedding model (required)
ollama pull qwen3:8b           # Preferred language model

Step 4: Configure JVM Arguments

Before starting the game, add the JVM argument:

-Daiplayer.llmMode=<provider>

Examples:

  • Local Ollama: -Daiplayer.llmMode=ollama
  • OpenAI: -Daiplayer.llmMode=openai
  • Google Gemini: -Daiplayer.llmMode=gemini

Step 5: Launch the Game

After starting Minecraft, use the command to create a bot:

/bot spawn <botName> <training|play>
  • training — training mode (without LLM connection)
  • play — normal mode with full AI capabilities

Key Features of Version 1.0.5.2

Custom OpenAI Provider Support

Now you can use alternative AI providers compatible with the OpenAI API standard, such as OpenRouter, TogetherAI, Perplexity, and others.

How to Use Custom Providers

  1. Set the system property: -Daiplayer.llmMode=custom
  2. Delete the existing settings.json5 file in the config folder
  3. Open the API key configuration screen in the game (/configMan)
  4. Set the base URL and API key of your provider
  5. Save settings and refresh the model list

Supported Providers

New Features in Version 1.0.5.1

Updated NLP System

Completely redesigned natural language processing system - no more "I didn't understand you" messages. This is an experimental system that continues to be improved.

RAG System with Web Search

New retrieval-augmented generation system integrated with database and web search. Now AI provides accurate information about Minecraft, significantly reducing errors.

Task Chaining System

Added a system for breaking down complex instructions into sequences of simple tasks. You give a high-level command → the bot automatically breaks it down into smaller tasks → executes step by step.

Supported Tasks:

  • Moving to a location
  • Mining resources at a specified location
  • Detecting nearby blocks and entities
  • Reporting statistics (health, oxygen, hunger, etc.)

Bug Fixes

  • Fixed issue with reading JVM arguments
  • Removed dependency on owo-lib, now uses own configuration system
  • Fixed issues with saving API keys
  • Added new launcher detection system
  • Updated Config Manager interface with search and responsive design

Future Plans

  • Combat and Survival Enhancements: Bot will use weapons to defend against mobs
  • Improved Navigation System: Smart movement through water and complex terrain
  • Self-Goal Assignment System: Bot will set goals for itself like a real player
  • Mood System: Adding emotional context and varied behavior
  • Player2 Integration: One of the most anticipated features

Important Notes

  • It's recommended to create a separate modpack for this mod, as Ollama must run in the background
  • Ollama consumes minimal memory and shouldn't significantly affect performance
  • The nomic-embed-text embedding model is always required for operation
Project members
shasankp000

shasankp000

Developer

Created: 23 Jul 2024

ID: 48621