MARL-DPP is an open-source framework enabling multi-agent reinforcement learning (MARL) with enforced diversity through Determinantal Point Processes (DPP). Traditional MARL approaches often suffer from policy convergence to similar behaviors; MARL-DPP addresses this by incorporating DPP-based measures to encourage agents to maintain diverse action distributions. The toolkit provides modular code for embedding DPP in training objectives, sampling policies, and managing exploration. It includes ready-to-use integration with standard OpenAI Gym environments and the Multi-Agent Particle Environment (MPE), along with utilities for hyperparameter management, logging, and visualization of diversity metrics. Researchers can evaluate the impact of diversity constraints on cooperative tasks, resource allocation, and competitive games. The extensible design supports custom environments and advanced algorithms, facilitating exploration of novel MARL-DPP variants.

Meta-Agent-with-More-Agents is an extensible open-source framework that implements a meta agent architecture allowing multiple specialized sub-agents to collaborate on complex tasks. It leverages LangChain for agent orchestration and OpenAI APIs for natural language processing. Developers can define custom agents for tasks like data extraction, sentiment analysis, decision-making, or content generation. The meta agent coordinates task decomposition, dispatches objectives to appropriate agents, gathers their outputs, and iteratively refines results via feedback loops. Its modular design supports parallel processing, logging, and error handling. Ideal for automating multi-step workflows, research pipelines, and dynamic decision support systems, it simplifies building robust distributed AI systems by abstracting inter-agent communication and lifecycle management.

MIU CS589 AI Agent API offers a standardized interface for building custom AI agents. Developers can define agent behaviors, integrate external tools or services, and handle streaming or batch responses via HTTP endpoints. The framework handles authentication, request routing, error handling and logging out of the box. It is fully extensible—users can register new tools, adjust agent memory, and configure LLM parameters. Suitable for experimentation, demos, and production prototypes, it simplifies multi-tool orchestration and accelerates AI agent development without locking you into a monolithic platform.

Modular LLM Architecture is designed to simplify the creation of customized LLM-driven applications through a composable, modular design. It provides core components such as memory modules for session state retention, tool interfaces for external API calls, prompt managers for template-based or dynamic prompt generation, and orchestration engines to control agent workflow. You can configure pipelines that chain together these modules, enabling complex behaviors like multi-step reasoning, context-aware responses, and integrated data retrieval. The framework supports multiple LLM backends, allowing you to switch or mix models, and offers extensibility points for adding new modules or custom logic. This architecture accelerates development by promoting reuse of components, while maintaining transparency and control over the agent’s behavior.

MultiAgentes provides a modular architecture for defining environments and agents, supporting synchronous and asynchronous multi-agent interactions. It includes base classes for environments and agents, predefined scenarios for cooperative and competitive tasks, tools for customizing reward functions, and APIs for agent communication and observation sharing. Visualization utilities allow real-time monitoring of agent behaviors, while logging modules record performance metrics for analysis. The framework integrates seamlessly with Gym-compatible reinforcement learning libraries, enabling users to train agents using existing algorithms. MultiAgentes is designed for extensibility, allowing developers to add new environment templates, agent types, and communication protocols to suit diverse research and educational use cases.

Notte is a developer-centric Python framework designed for orchestrating AI agents powered by large language models. It provides built-in memory modules to store and retrieve conversational context, flexible tool integration for external APIs or custom functions, and a planning engine that sequences tasks. With Notte, you can rapidly prototype conversational assistants, data analysis bots, or automated workflows, while benefiting from open-source extensibility and cross-platform support.

rag-services is an extensible platform that breaks down RAG pipelines into discrete microservices. It offers a document store service, a vector index service, an embedder service, multiple LLM inference services, and an orchestrator service to coordinate workflows. Each component exposes REST APIs, allowing you to mix and match databases and model providers. With Docker and Docker Compose support, you can deploy locally or in Kubernetes clusters. The framework enables scalable, fault-tolerant RAG solutions for chatbots, knowledge bases, and automated document Q&A.

SonAgent is an extensible open-source framework designed for building, organizing, and running AI agents in Python. It provides core modules for memory storage, tool wrappers, planning logic, and asynchronous event handling. Developers can register custom tools, integrate language models, manage long-term agent memory, and orchestrate multiple agents to collaborate on complex tasks. SonAgent’s modular design accelerates the development of conversational bots, workflow automations, and distributed agent systems.

EspressoAI provides developers with a structured environment to design, configure, and deploy AI agents powered by large language models. It supports tool registration and invocation from within agent workflows, manages conversational context via built-in memory modules, and allows chaining of prompts for multi-step reasoning. Developers can integrate external APIs, custom plugins, and conditional logic to tailor agent behavior. The framework’s modular design ensures extensibility, enabling teams to swap components, add new capabilities, or adapt to proprietary LLMs without rewriting core logic.

Huginn is a versatile, open-source automation framework that lets users create agents to monitor, gather, and act on data from various sources such as websites, APIs, social media, and email. Each agent can be configured to trigger on events, transform data, and pass it to other agents or external services. With built-in scheduling, logging, and a rich library of agent types—like RSSAgent, EmailAgent, WebhookAgent, and DataOutputAgent—Huginn supports complex workflows and conditional logic. It runs on Linux, macOS, Windows, or Docker, and can be extended with custom Ruby code or Docker containers for specialized tasks and integrations.

Llama-Agent is a developer-focused toolkit for creating intelligent AI agents powered by large language models. It offers tool integration to call external APIs or functions, memory management to store and retrieve context, and chain-of-thought planning to break down complex tasks. Agents can execute actions, interact with custom environments, and adapt through a plugin system. As an open-source project, it supports easy extension of core components, enabling rapid experimentation and deployment of automated workflows across various domains.

Query-Bot integrates document ingestion, text chunking, and vector embeddings to build a searchable index from PDFs, text files, and Word documents. Using LangChain and OpenAI GPT-3.5 Turbo, it processes user queries by retrieving relevant document passages and generating concise answers. The Streamlit-based UI allows users to upload files, track conversation history, and adjust settings. It can be deployed locally or on cloud environments, offering an extensible framework for custom agents and knowledge bases.

FastAPI Agents provides a robust service layer for developing LLM-based agents using the FastAPI web framework. It allows you to define agent behaviors with LangChain chains, tools, and memory systems. Each agent can be exposed as a standard REST endpoint, supporting asynchronous requests, streaming responses, and customizable payloads. Integration with vector stores enables retrieval-augmented generation for knowledge-driven applications. The framework includes built-in logging, monitoring hooks, and Docker support for containerized deployment. You can easily extend agents with new tools, middleware, and authentication. FastAPI Agents accelerates the production readiness of AI solutions, ensuring security, scalability, and maintainability of agent-based applications in enterprise and research settings.

AI-Agents provides a modular toolkit to create autonomous AI agents capable of task planning, execution, and self-monitoring. It offers built-in support for tool integration—such as web search, data processing, and custom APIs—and features a memory component to retain and recall context across interactions. With a flexible plugin system, agents can dynamically load new capabilities, while asynchronous execution ensures efficient multi-step workflows. The framework leverages LangChain for advanced chain-of-thought reasoning and simplifies deployment in Python environments on macOS, Windows, or Linux.

Advanced_RAG provides a modular pipeline for retrieval-augmented generation tasks, including document loaders, vector index builders, and chain managers. Users can configure different vector databases (FAISS, Pinecone), customize retriever strategies (similarity search, hybrid search), and plug in any LLM to generate contextual answers. It also supports evaluation metrics and logging for performance tuning and is designed for scalability and extensibility in production environments.

Agent Adapters is designed to provide developers with a consistent interface for connecting AI agents to external services and frameworks. Through its pluggable adapter architecture, it offers prebuilt adapters for HTTP APIs, messaging platforms like Slack and Teams, and custom tool endpoints. Each adapter handles request parsing, response mapping, error handling, and optional logging or monitoring hooks. Developers can also register custom adapters by implementing a defined interface and configuring adapter parameters in their agent settings. This streamlined approach reduces boilerplate code, ensures uniform workflow execution, and accelerates the deployment of agents across multiple environments without rewriting integration logic.

Agentle provides a structured framework for developers to build custom AI agents with minimal boilerplate. It supports defining agent workflows as sequences of tasks, seamless integration with external APIs and tools, conversational memory management for context preservation, and built-in logging for auditability. The library also offers plugin hooks to extend functionality, multi-agent coordination for complex pipelines, and a unified interface to run agents locally or deploy via HTTP APIs.

AgentMesh provides a modular infrastructure for developers to create networks of AI agents, each focusing on a specific task or domain. Agents can be dynamically discovered and registered at runtime, exchange messages asynchronously, and follow configurable routing rules. The framework handles retries, fallbacks, and error recovery, allowing multi-agent pipelines for data processing, decision support, or conversational use cases. It integrates easily with existing LLMs and custom models via a simple plugin interface.

AgentUniverse provides a unified Python SDK to design, orchestrate, and run autonomous AI agents. Developers can define agent behaviors, integrate external tools or APIs, maintain conversational memory, and sequence multi-step tasks. Supporting LangChain, custom tool plugins, and configurable runtime environments, it accelerates agent development and deployment. Built-in monitoring and logging enable real-time insights, while its modular architecture allows easy extension with new capabilities or AI models.

Crayon empowers developers to build autonomous AI agents in JavaScript/Node.js that can call external APIs, maintain conversation history, plan multi-step tasks, and handle asynchronous processes. At its core, Crayon implements a planning-execution loop that breaks down high-level goals into discrete actions, integrates with custom toolkits, and utilizes memory modules to store and recall information across sessions. The framework supports multiple memory backends, plugin-based tool integration, and comprehensive logging for debugging. Developers can configure agent behavior through prompts and YAML-based pipelines, enabling complex workflows like data scraping, report generation, and interactive chatbots. Crayon's architecture promotes extensibility, allowing teams to integrate domain-specific tools and tailor agents to unique business requirements.