A fully open source framework for creating RL training swarms over the internet.

为 Agent Swarm 打造的 IDE ，支持 Kimi-2.5, GLM-4.7 等（即使不像 K2.5 那样经过强化学习的模型也能用）An IDE built for Agent Swarms, supporting Kimi-2.5, GLM-4.7, and more (compatible with models even without RL-reinforcement like K2.5).

Simple and well detailed guide on how to run rl-swarm

Detailed Guide on How to Contribute to Gensyn RL-Swarm

Step By Step Guide, How u can run the Gensyn-ai/Rl-Swarm Node ~ (Mac/linux)

JAX and PZ RL envs + algorithms for swarms of CrazyFlies

No description available

Training of Drone Swarms using StableBaselines3, PettingZoo, AirSim and UE4

This ROS kinectic workspace src folder, which was created on Ubuntu 16.04. Here I worked on ROS1 projects like 2d & 3D SLAM, Motion Planning, SWARM of drone, RL based drone, Surveilling Robot etc.

Multi agent reinforcement learning for intelligent active matter

Smart contracts for the RL Swarm project, focusing on coordinating swarm behavior onchain.

2026 swarm Agent 年，swarm Agent 、Agent team、 ai coding、skill、memory、evolve、agentic RL 等 AI Agent集合

No description available

Technical report for the RL Swarm framework - a fully open source framework for creating RL training swarms over the internet.

Flock and swarm multi-agent RL training environments implemented in JAX

Using reinforcement learning algorithm for multi-agent target assignment

Training of Drone Swarms using StableBaselines3, PettingZoo, AirSim and UE4

This repository contains a jupyter notebook which includes the code for Q-learning and SARSA based path planning of 2 UAVs in 2D grid environment with three different sizes: 6X6, 8X8, and 10X10.

No description available

Gensyn RL-Swarm on QuickPod

No description available

Reinforcement Learning environment for swarm rescue problem

A ROS2 Reinforcement Learning Project for Robot Swarm Navigation

Reinforcement learning to make colloids smart

No description available

Using Deep Reinforcement Learning (associated with Deep Learning) to control a swarm of drones for dynamic area maximization problem

GensynAi(Testnet). A fully open source framework for creating RL training swarms over the internet.

This project introduces the Model Context Protocol (MCP), a lightweight communication layer enabling UAV swarms in disaster zones to share real-time context. Each UAV, guided by RL agents, makes decentralized decisions for coverage, energy, and reliability. A digital twin simulation shows efficiency gains over context-agnostic swarms.

rl-swarm-mac-stable

No description available