Game Playing Reinforcement Learning ML Frameworks

RL agents trained to play video games (Atari, Mario, DOOM, etc.) using deep Q-learning, policy gradients, and other RL algorithms. Does NOT include general RL frameworks, robotics control, or non-game environments.

There are 34 game playing reinforcement learning frameworks tracked. 5 score above 50 (established tier). The highest-rated is Talendar/flappy-bird-gym at 64/100 with 132 stars and 93 monthly downloads.

Get all 34 projects as JSON

curl "https://pt-edge.onrender.com/api/v1/datasets/quality?domain=ml-frameworks&subcategory=game-playing-reinforcement-learning&limit=20"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.

#	Framework	Score	Tier	Stars	Language
1	Talendar/flappy-bird-gym An OpenAI Gym environment for the Flappy Bird game	64	Established	132	Python
2	Farama-Foundation/ViZDoom Reinforcement Learning environments based on the 1993 game Doom :godmode:	53	Established	1,990	C++
3	aleju/mario-ai Playing Mario with Deep Reinforcement Learning	51	Established	695	Lua
4	chris-chris/pysc2-examples StarCraft II - pysc2 Deep Reinforcement Learning Examples	51	Established	756	Python
5	gavento/gamegym A game theory framework with examples and algorithms	50	Established	74	Python
6	Kaixhin/Atari Persistent advantage learning dueling double DQN for the Arcade Learning Environment	49	Emerging	263	Lua
7	gsurma/atari AI research environment for the Atari 2600 games 🤖.	49	Emerging	263	Python
8	uber-research/atari-model-zoo A binary release of trained deep reinforcement learning models trained in...	45	Emerging	202	Jupyter Notebook
9	SeanCole02/doom-neuron Human brain cells play Doom (CL1)	45	Emerging	185	Python
10	matlab-deep-learning/playing-Pong-with-deep-reinforcement-learning Train a reinforcement learning agent to play a variation of Pong®	43	Emerging	46	MATLAB
11	thinkingparticle/deep_rl_pong_keras Deep Reinforcement Learning Policy Gradients Method - Pong game - Keras	32	Emerging	22	Jupyter Notebook
12	nikp06/icyAI Gym for training an AI to learn Icy Tower through neuroevolution	31	Emerging	37	Python
13	HugoTini/GymGodot Reinforcement learning with Godot	30	Emerging	28	GDScript
14	slrbl/reinforcement-learning-game A random environment reinforcement learning-powered Mario game	27	Experimental	9	JavaScript
15	tims457/Super-Mario-Land-RL Super Mario Land Reinforcement Learning with Pufferlib	25	Experimental	2	Jupyter Notebook
16	Lollorm/Super-Mario-AI Training AI Agents to Beat Super Mario Bros and Super Mario World Using...	25	Experimental	3	Python
17	gandalf1819/Reinforcement-Learning-Comparative-Study Comparative study of Reinforcement Learning Algorithms on Ping Pong game: In...	24	Experimental	2	Jupyter Notebook
18	ThekingGST/DQN-Atari Train an AI to play retro Atari games! This project implements DeepMind's...	23	Experimental	1	Python
19	WillianFuks/Pong Implementation of Reinforcement Learning Policy Gradients algorithms to play...	23	Experimental	2	Python
20	shreyashreddy/Block-Reign 🎮 Experience grid battles against a self-learning AI that improves from each...	22	Experimental	—	Python
21	akndr05/RB2 Recompile and run Dragon Ball Raging Blast 2 using ReXGlue SDK tools for...	22	Experimental	—	C++
22	ChernyakKonstantin/godot_gym_api Godot Gym API is an Open Source framework for using Godot3 game engine as...	20	Experimental	28	GDScript
23	primaryobjects/qlearning A game using Q-Learning artificial intelligence.	17	Experimental	1	JavaScript
24	WillKirkmanM/mario-party-finder 18 Layered Convolutional Neural Network to Predict Any Mario Party Minigame...	17	Experimental	5	Python
25	ai-creature/ai-creature.github.io The AI Creature is a Reinforcement Learning agent based on the Soft-Actor...	17	Experimental	3	JavaScript
26	Laggg/neural-env-surviv Train environment model for RL based agent in browser-based multiplayer...	16	Experimental	26	Jupyter Notebook
27	DavideEspositoPelella/SuperMario-RL Application of the DDQN with ICM algorithm to the SuperMario Game...	15	Experimental	2	Python
28	pstepanovum/browser-marl-hideseek Multi-agent reinforcement learning framework for training NPCs in...	15	Experimental	—	JavaScript
29	vanillaSky00/neural-arcade 3 AI agents that learn to play Arkanoid, Swimming Squid, and TankMan, from...	14	Experimental	—	Python
30	mharrell/BreakoutBot Building a deep reinforcement learning agent using PPO and PyTorch to train...	14	Experimental	—	Python
31	SicrediRC/FreeHoopRL 🏀 Train an AI agent to master basketball shooting using a deep Q-network in...	14	Experimental	—	Python
32	RobinKa/warlock-rl Reinforcement learning for Warlock Brawl, and an ECS implementation of...	11	Experimental	—	TypeScript
33	cudnah124/Q-bert DQN Agent for Q*bert: Compare vanilla, double & dueling deep Q-networks on ...	11	Experimental	—	Python
34	VictorNarov/QLearning-GVGAI80 QLearning algorithm of Reinforcement Learning implemented in the GVGAI...	10	Experimental	1	Java

Comparisons in this category

deep_rl_pong_keras and Pong (32 vs 23)