CybMASDE enables researchers and developers to build, configure, and execute multi-agent simulations with deep reinforcement learning. Users can author custom scenarios, define agent roles and reward functions, and plug in standard or custom RL algorithms. The framework includes environment servers, networked agent interfaces, data collectors, and rendering utilities. It supports parallel training, real-time monitoring, and model checkpointing. CybMASDE’s modular architecture allows seamless integration of new agents, observation spaces, and training strategies, accelerating experimentation in cooperative control, swarm behavior, resource allocation, and other multi-agent use cases.

Eliza is a lightweight, open-source conversational framework that simulates a psychotherapist via pattern matching and scripted templates. Developers can define custom scripts, patterns, and memory variables to tailor responses and conversation flows. It runs in any modern browser or webview environment, supports multiple sessions, and logs interactions for analysis. Its extensible architecture allows integration into web pages, mobile apps, or desktop wrappers, making it a versatile tool for education, research, prototype development, and interactive installations.

SwarmZero offers a scalable, open-source environment for defining, managing, and executing swarms of AI agents. Developers can declare agent roles, customize prompts, and chain workflows via a unified Orchestrator API. The framework integrates with major LLM providers, supports plugin extensions, and logs session data for debugging and performance analysis. Whether coordinating research bots, content creators, or data analyzers, SwarmZero streamlines multi-agent collaboration and ensures transparent, reproducible results.

RAGENT is designed to create autonomous AI agents that can interact with diverse tools and data sources. Under the hood, it uses retrieval-augmented generation to fetch relevant context from local files or external sources and then composes responses via OpenAI models. Developers can plug in tools for web search, browser automation with Selenium, file read/write operations, code execution in secure sandboxes, and OCR for image text extraction. The framework manages conversation memory, handles tool orchestration, and supports custom prompt templates. With RAGENT, teams can rapidly prototype intelligent agents for document Q&A, research automation, content summarization, and end-to-end workflow automation, all within a Python environment.

Lagent is a developer-focused framework that enables creation of intelligent agents on top of large language models. It offers dynamic planning modules that break tasks into subgoals, memory stores to maintain context over long sessions, and tool integration interfaces for API calls or external service access. With customizable pipelines, users define agent behaviors, prompting strategies, error handling, and output parsing. Lagent’s logging and debugging tools help monitor decision steps, while its scalable architecture supports local, cloud, or enterprise deployments. It accelerates building autonomous assistants, data analysers, and workflow automations.

LaVague is an open-source framework designed for building and deploying web agents quickly and efficiently. Users can create various agents that automate tasks across web applications, from data entry to comprehensive information retrieval. The framework supports integration with local models, such as Llama 3 8b, making it a versatile choice for enterprises looking to enhance their operations with AI-driven automation. With LaVague, developers can adapt agents to fit specific workflows, enhancing productivity and efficiency.

Linguistic Agent System is an open-source Python framework designed for constructing intelligent agents that leverage language models to plan and execute tasks. It includes components for memory management, tool registry, planner, and executor, allowing agents to maintain context, call external APIs, perform web searches, and automate workflows. Configurable via YAML, it supports multiple LLM providers, enabling rapid prototyping of chatbots, content summarizers, and autonomous assistants. Developers can extend functionality by creating custom tools and memory backends, deploying agents locally or on servers.

LLM-Powered RAG System is a developer-focused framework for building retrieval-augmented generation (RAG) pipelines. It provides modules for embedding document collections, indexing via FAISS, Pinecone, or Weaviate, and retrieving relevant context at runtime. The system uses LangChain wrappers to orchestrate LLM calls, supports prompt templates, streaming responses, and multi-vector store adapters. It simplifies end-to-end RAG deployment for knowledge bases, allowing customization at each stage—from embedding model configuration to prompt design and result post-processing.

llm-lab provides a flexible toolkit for creating intelligent agents using large language models. It includes an agent orchestration engine, support for custom prompt templates, memory and state tracking, and seamless integration with external APIs and plugins. Users can write scenarios, define toolchains, simulate interactions, and collect performance logs. The framework also offers a built-in testing suite to validate agent behavior against expected outcomes. Extensible by design, llm-lab enables developers to swap LLM providers, add new tools, and evolve agent logic through iterative experimentation.

MAPF_G2RL is an open-source research framework that bridges graph theory and deep reinforcement learning to tackle the multi-agent path finding (MAPF) problem. It encodes nodes and edges into vector representations, defines spatial and collision-aware reward functions, and supports various RL algorithms such as DQN, PPO, and A2C. The framework automates scenario creation by generating random graphs or importing real-world maps, and orchestrates training loops that optimize policies for multiple agents simultaneously. After learning, agents are evaluated in simulated environments to measure path optimality, makespan, and success rates. Its modular design allows researchers to extend core components, integrate new MARL techniques, and benchmark against classical solvers.

MiniAgent is a minimalistic open-source framework built in Python for constructing autonomous AI agents capable of planning and executing complex workflows. At its core, MiniAgent includes a task planning module that decomposes high-level goals into ordered steps, an execution controller that runs each step sequentially, and built-in adapters for integrating external tools and APIs, including web services, databases, and custom scripts. It also features a lightweight memory management system to persist conversational or task context. Developers can easily register custom action plugins, define policy rules for decision-making, and extend tool functionality. With support for OpenAI models and local LLMs, MiniAgent enables rapid prototyping of chatbots, digital workers, and automated pipelines, all under an MIT license.

mini-AlphaStar demystifies the complex AlphaStar architecture by offering an accessible, open-source PyTorch framework for StarCraft II AI development. It features spatial feature encoders for screen and minimap inputs, non-spatial feature processing, LSTM memory modules, and separate policy and value networks for action selection and state evaluation. Using imitation learning to bootstrap and reinforcement learning with self-play for fine-tuning, it supports environment wrappers compatible with StarCraft II via pysc2, logging through TensorBoard, and configurable hyperparameters. Researchers and students can generate datasets from human gameplay, train models on custom scenarios, evaluate agent performance, and visualize learning curves. The modular codebase enables easy experimentation with network variants, training schedules, and multi-agent setups. Designed for education and prototyping rather than production deployment.

Multi-Agent Systems provides a comprehensive toolkit for creating, controlling, and observing interactions among autonomous agents. Developers can define agent classes with custom decision-making logic, set up complex environments with configurable resources and rules, and implement communication channels for information exchange. The framework supports synchronous and asynchronous scheduling, event-driven behaviors, and integrates logging for performance metrics. Users can extend core modules or integrate external AI models to enhance agent intelligence. Visualization tools render simulations in real-time or post-process, helping analyze emergent behaviors and optimize system parameters. From academic research to prototype distributed applications, Multi-Agent Systems simplifies end-to-end multi-agent simulations.

This open-source framework combines the AgentSimJs agent modeling library with Three.js's 3D graphics engine to deliver interactive, browser-based multi-agent simulations. Users can define agent types, behaviors, and environmental rules, configure collision detection and event handling, and visualize simulations in real time with customizable rendering options. The library supports dynamic controls, scene management, and performance tuning, making it ideal for research, education, and prototyping of complex agent-based scenarios.

Multi-Agent Architecture provides a scalable, extensible platform to define, register, and coordinate multiple AI agents working together on a shared objective. It includes a message broker, lifecycle management, dynamic agent spawning, and customizable communication protocols. Developers can build specialized agents (e.g., data fetchers, NLP processors, decision-makers) and plug them into the core runtime to handle tasks ranging from data aggregation to autonomous decision workflows. The framework’s modular design supports plugin extensions and integrates with existing ML models or APIs.

Multi-Agent ColComp is an extensible, open-source framework for orchestrating a team of AI agents to work together on complex tasks. Developers can define distinct agent roles, configure communication channels, and share contextual data through a unified memory store. The library includes plug-and-play components for negotiation, coordination, and consensus building. Example setups demonstrate collaborative text generation, distributed planning, and multi-agent simulation. Its modular design supports easy extension, enabling teams to prototype and evaluate multi-agent strategies rapidly in research or production environments.

MultiAgentPacman offers a Python-based game environment where users can implement, visualize, and benchmark multiple AI agents in the Pacman domain. It supports adversarial search algorithms like minimax, expectimax, alpha-beta pruning, as well as custom reinforcement learning or heuristic-based agents. The framework includes a simple GUI, command-line controls, and utilities to log game statistics and compare agent performance under competitive or cooperative scenarios.

MultiAgentPuzzleSolver provides a modular environment where independent AI agents work together to solve puzzles such as sliding tiles, Rubik’s Cube, and logic grids. Agents share state information, negotiate subtask assignments, and apply diverse heuristics to explore the solution space more effectively than single-agent approaches. Developers can plug in new agent behaviors, customize communication protocols, and add novel puzzle definitions. The framework includes tools for real-time visualization of agent interactions, performance metrics collection, and experiment scripting. It supports Python 3.8+, standard libraries, and popular ML toolkits for seamless integration into research projects.

OpenAI Agents MCP Server provides a robust foundation for deploying and managing autonomous agents powered by OpenAI models. It exposes a flexible RESTful API to create, configure, and control agents, enabling developers to orchestrate multi-step tasks, coordinate interactions between agents, and maintain persistent memory across sessions. The framework supports plugin-like tool integrations, advanced conversation logging, and customizable planning strategies. By abstracting infrastructure concerns, MCP Server streamlines the development pipeline, facilitating rapid prototyping and scalable deployment of conversational assistants, workflow automations, and AI-driven digital workers in production environments.

RoboCup Rescue Agent Simulation is an open-source framework that models urban disaster environments where multiple AI-driven agents collaborate to locate and rescue victims. It offers interfaces for navigation, mapping, communication, and sensor integration. Users can script custom agent strategies, run batch experiments, and visualize agent performance metrics. The platform supports scenario configuration, logging, and result analysis to accelerate research in multi-agent systems and disaster response algorithms.