Open-Source AI Agent Frameworks

Pydantic AI is a Python SDK for developers building agents into their applications. It uses Pydantic validation to give application code structured results and check the arguments agents pass to tools.

Swarm Fleet manages multiple terminal AI agents on your machine for developers coordinating coding, review and ticket work. Each agent has its own terminal, with shared controls for supervising their work.

Langroid is a Python framework for developers building applications where AI agents divide up a task. Each agent can maintain its own conversation and use documents or tools, then delegate work to other agents.

Inspect AI is a Python framework for researchers and developers evaluating language models and AI agents. It measures coding, reasoning, knowledge, behavior and multimodal understanding, including tasks where agents must take actions to succeed.

MetaGPT is a Python AI agent framework for developers who want to generate software projects using local models through Ollama or remote model APIs. The framework runs on your machine and writes a project repository locally. Its example configuration uses OpenAI, which sends model requests to that provider and requires API credentials.

Semantic Kernel is an SDK for developers building AI agents and coordinated workflows inside applications. It lets developers choose local model backends or cloud providers for those applications.

Rasa Open Source is a Python framework for developers building text and voice assistants on their own servers. It provides natural-language understanding, dialogue management and integrations with messaging channels.

NullClaw runs AI agents on your own hardware or server for coding work and automated background tasks. It's aimed at developers who want an assistant that can carry out multi-step jobs with controls over its access and actions.

AgentScope is a Python framework for developers building AI agent applications on their own servers. It includes an SDK, a FastAPI backend and a web UI for applications with multiple users and separate sessions.

PenguinHarness builds runnable AI agent applications from plain-language requirements on your computer or server. It's for developers who want agents to handle app creation, evaluation, tuning and deployment.

BeeAI Framework is for developers building AI agents and systems where several agents work together. It lets developers enforce rules on agent behavior while retaining the model's ability to reason.

TEN Framework is a self-hosted framework for developers building voice agents and multimodal conversational applications. Its default assistant example depends on external services even when the application runs on your hardware.

Deep Agents is an AI agent framework for developers building assistants that carry out multi-step work. It provides planning and task execution defaults that you can customize without forking the project.

II Agent is an AI agent for developers and teams who want an assistant to carry out research, build applications and handle documents. It can act through a browser, shell and files to complete tasks that need several steps.

MartinLoop runs around coding agents to keep software jobs within agreed boundaries and check their results. It's for developers and engineering teams who want to delegate longer tasks while retaining limits, recovery information and a record of what happened.

Pi Extensible Workflows is an extension for developers who want repeatable processes involving several AI agents. It coordinates resumable workflows inside the Pi host environment and lets you inspect work while it runs.

OWL is a Python framework built on CAMEL-AI for developers who want multiple AI agents to work together on tasks. It combines research, file analysis, and actions in a browser or terminal within one task workflow.

Rivet is a desktop visual AI agent builder for developers working with connected prompts and application logic. Teams can inspect complex LLM workflows as graphs and run those graphs inside their own applications.

GPTeam is a self-hosted AI agent simulation for developers and researchers studying how agents communicate and work toward assigned goals. It runs locally in Python, while configured providers handle model calls through cloud APIs. The code uses the MIT license.

Qwen-Agent is a Python framework for developers building assistants that use Qwen models to read files and carry out tasks with tools. It provides ready-made agent components and example applications for custom assistants and browser assistance.

ChatDev is a self-hosted AI agent platform for people building research, data analysis and software workflows. Its DevAll application runs the workflow backend on your computer or server, with a browser console for designing and overseeing agent teams. Model access uses your chosen provider's API credentials and endpoint; choosing a cloud provider sends model requests to that service.

OpenFang runs recurring AI tasks on your own machine and reports their results to a desktop dashboard. It's for people who want research, monitoring or business automation to continue on a schedule without starting each job in chat.

AutoGPT is an AI agent platform for teams automating recurring content, sales, research and operations work. Its hosted service centers on Experts coordinated by Otto; self-hosting provides the core workflow builder and agent runtime.

VibeHQ Hub coordinates coding agents for developers who want several agents to work on the same project without conflicting interfaces or unfinished handoffs. It adds shared coordination to Claude Code, Codex and Gemini CLI while preserving each agent's native capabilities.

FinRobot is an Apache 2.0 open-source equity research system for analysts and quantitative researchers. Its desktop app runs financial calculations locally and uses an LLM to interpret results and write reports. Built-in OpenAI and Anthropic connections send model requests to those providers; custom OpenAI-compatible connections can use a local vLLM backend. Research also needs external market data.

Mirascope is an open-source development library for building AI applications with local models or cloud providers. Its documented local backends include Ollama, vLLM and LM Studio, plus direct inference through MLX on Apple Silicon Macs. Developers can use a shared model interface while keeping inference on their own hardware or sending requests to OpenAI, Anthropic or Google.

mistral.rs is an inference engine for developers who want to run AI models on their own computer or server. It supports browser chat and inference embedded directly in Python or Rust applications.

npcpy is a Python library for developers and researchers building AI agents and applications with language models. It supports local LLM backends as well as cloud providers, so projects can choose where model requests run.

Ouroboros is a locally installed workflow engine for developers who want to define requirements and check an AI coding agent's output. It runs through a terminal CLI, a Claude Code plugin or an MCP server. Your chosen coding runtime supplies the agent and model access. When used with Claude through Claude Code's standard cloud connection, model requests go to that external service.

AgentField is a self-hosted AI agent platform for developers building services that need to coordinate tasks under shared access rules. It combines execution, identity and audit records in infrastructure you control.

BabyAGI is a self-hosted Python framework for experienced developers exploring AI agent development. Its local application manages reusable code and provides a browser dashboard. Functions can depend on external services and API keys, including an OpenAI API key.

quackd lets robot builders give language-based goals to robots with existing movement and manipulation skills. It runs the planning on a computer while the robot's own controllers carry out the actions. Real-hardware use covers the LeRobot SO-101 arm; the other adapters have only run in simulators or mocks.

LangChain is an open-source framework for developers building AI agents and LLM applications, including apps that run models locally through Ollama. Its Python ChatOllama integration connects to open-weight models such as gpt-oss. You write the application in code, while Ollama handles local inference.

Strands Harness SDK is an open source toolkit for developers building AI agents who need control over their behavior. It handles the agent loop while letting you choose its components.