mario-ai is an open-source Python framework that leverages NeuroEvolution of Augmenting Topologies (NEAT) to evolve neural network-based AI agents capable of playing Super Mario Bros. It integrates with OpenAI Gym’s SuperMario environment, providing customizable fitness functions, real-time training visualization, genome saving/loading, and performance monitoring. Users can adjust mutation parameters, define evaluation metrics, and visualize learned network topologies.
AI Platforms & Frameworks
Gaming
#neuroevolution
#AI gaming
#Super Mario
#OpenAI Gym
#genetic algorithms
#Real-time Visualization
#custom fitness functions
#machine learning in games
#AI agent development
#gaming AI tools
#Performance Tracking
#Open-source AI
#AI model training
#evolutionary strategies
#game intelligence
#neural network training
#educational AI tools
#hyperparameter optimization
#evolving neural networks
#AI benchmarking
...
Promote this Tool
Update this Tool
mario-ai
mario-ai is an open-source Python framework that leverages NeuroEvolution of Augmenting Topologies (NEAT) to evolve neural network-based AI agents capable of playing Super Mario Bros. It integrates with OpenAI Gym’s SuperMario environment, providing customizable fitness functions, real-time training visualization, genome saving/loading, and performance monitoring. Users can adjust mutation parameters, define evaluation metrics, and visualize learned network topologies.
AI Platforms & Frameworks
Gaming
#neuroevolution
#AI gaming
#Super Mario
#OpenAI Gym
#genetic algorithms
#Real-time Visualization
#custom fitness functions
#machine learning in games
#AI agent development
#gaming AI tools
#Performance Tracking
#Open-source AI
#AI model training
#evolutionary strategies
#game intelligence
#neural network training
#educational AI tools
#hyperparameter optimization
#evolving neural networks
#AI benchmarking
...
Featured
What is mario-ai?
The mario-ai project offers a comprehensive pipeline for developing AI agents to master Super Mario Bros. using neuroevolution. By integrating a Python-based NEAT implementation with the OpenAI Gym SuperMario environment, it allows users to define custom fitness criteria, mutation rates, and network topologies. During training, the framework evaluates generations of neural networks, selects high-performing genomes, and provides real-time visualization of both gameplay and network evolution. Additionally, it supports saving and loading trained models, exporting champion genomes, and generating detailed performance logs. Researchers, educators, and hobbyists can extend the codebase to other game environments, experiment with evolutionary strategies, and benchmark AI learning progress across different levels.
Who will use mario-ai?
- AI researchers
- Game AI hobbyists
- Educational instructors
- Students in AI
- Evolutionary algorithm enthusiasts
How to use the mario-ai?
- Step1: Clone the mario-ai repository from GitHub.
- Step2: Install dependencies with pip install -r requirements.txt and gym-super-mario-bros.
- Step3: Configure the NEAT parameters in config-feedforward.txt to adjust fitness and mutation settings.
- Step4: Run the training script (python train.py) to start evolving AI agents.
- Step5: Monitor training metrics and visualize neural networks via provided scripts.
- Step6: Once satisfied, save the best genome and use python play.py to watch the AI play.
Platform
- mac
- windows
- linux
mario-ai's Core Features & Benefits
The Core Features
- Neuroevolution via NEAT
- OpenAI Gym SuperMario integration
- Customizable fitness functions
- Real-time training visualization
- Save/load genome models
- Performance logging and export
The Benefits
- Automated game playing AI
- Highly customizable evolution parameters
- Educational tool for neuroevolution
- Open-source and extensible
- Easy model evaluation and visualization
mario-ai's Main Use Cases & Applications
- Educational demos on neuroevolution
- Research in evolutionary game AI
- Benchmarking AI performance
- Creating game-playing bots for entertainment
- Experimenting with custom fitness functions
FAQs of mario-ai
mario-ai Company Information
- aleju
mario-ai Reviews
5/5
mario-ai's Main Competitors and alternatives?
- marI/O by SethBling
- NEAT-Python
- OpenAI Gym Retro
You may also like:
Conversational AI assistant automating research, browser tasks, web scraping, and file management through natural language.
AI design agent automates workflow creating images, videos, 3D models up to 10x faster.
Short Circuit is a premier ChatGPT app for iPhone, iPad, and Mac.
SJinn is an AI-powered agent creating image, video, audio, and 3D content from descriptions.
Skywork AI is an innovative tool to enhance productivity using AI.
Refly.AI empowers non-technical creators to automate workflows using natural language and a visual canvas.
Eigent is an open-source AI workforce platform managing complex workflows via multi-agent collaboration.
A multi-agent reinforcement learning platform offering customizable supply chain simulation environments to train and evaluate AI agents effectively.
Airkit.ai is an AI agent that automates customer interactions and enhances communication channels.
Boostimize AI enhances e-commerce growth using personalized recommendations.
aiLEADS is an AI-powered lead generation agent designed to optimize sales processes.
Harmony is an AI Agent for streamlining coworking space management and enhancing community interactions.
Flowith is a canvas-based agentic workspace which offers free 🍌Nano Banana Pro and other effective models...
An AI agent that uses Minimax and Monte Carlo Tree Search to optimize tile placement and scoring in Azul.
AGM: AI Game Maker enables seamless game development with AI support.
An RL-based AI agent that learns optimal betting strategies to play heads-up limit Texas Hold'em poker efficiently.
An open-source reinforcement learning agent using PPO to train and play StarCraft II via DeepMind's PySC2 environment.
Open-source framework enabling implementation and evaluation of multi-agent AI strategies in a classic Pacman game environment.
BomberManAI is a Python-based AI agent that autonomously navigates and battles in Bomberman game environments using search algorithms.
SoccerAgent uses multi-agent reinforcement learning to train AI players for realistic soccer simulations and strategy optimization.
An AI-powered Dungeon Master that uses LLMs to generate dynamic D&D narrative, quests, and encounters in real-time.
FixArt AI offers free, unrestricted AI tools for image and video generation without sign-up.
An open-source multi-agent reinforcement learning framework enabling raw-level agent control and coordination in StarCraft II via PySC2.
An open-source RL agent for Yu-Gi-Oh duels, providing environment simulation, policy training, and strategy optimization.
PyGame Learning Environment provides a collection of Pygame-based RL environments for training and evaluating AI agents in classic games.
BotPlayers is an open-source framework enabling creation, testing, and deployment of AI game-playing agents with reinforcement learning support.
Gomoku Battle is a Python framework enabling developers to build, test, and pit AI agents in Gomoku games.
A multi-agent football simulation using JADE, where AI agents coordinate to compete in soccer matches autonomously.
FemaleSwitch is an AI-powered game that enhances female character experiences.
An AI agent that plays Pentago Swap by evaluating board states and selecting optimal placements using Monte Carlo Tree Search.
Samsung Ballie is a mobile AI assistant that monitors and interacts in your home.
AIpacman is a Python framework providing search-based, adversarial, and reinforcement learning agents to master the Pac-Man game.
All-in-one AI video creation studio that turns any text and images into full videos up to 30 minutes.