MetaGPT organizes AI agents into a software development team, with roles for product management, architecture, project management and engineering. It's a self-hosted Python framework for developers who want to generate software projects from plain-language requirements or build their own collaborative agents.
A request can produce a repository alongside user stories, competitive analysis and requirements documents. The agents also work on data structures, API designs and other project documentation. Their collaboration follows defined procedures modeled on software team workflows, giving each role a specific part of the work rather than treating the whole project as a single coding prompt.
The framework runs locally or in Docker. It supports Ollama as well as cloud backends including OpenAI, Azure and Groq; GPT-4-turbo and GPT-3.5-turbo appear among its model choices. Where model requests go depends on the backend you choose: Ollama supports local model use, while cloud API backends send requests to the selected provider.
MetaGPT is open source under the MIT license and works both as a command-line tool and a Python library. The library suits developers who want to incorporate agent teams into their own software. Its Data Interpreter can write code for data analysis and produce plots, including when used in a Jupyter notebook.
The documented setup requires Python 3.9 through 3.11, plus Node.js and pnpm.
Claim this page and we'll verify you by hand. MetaGPT gets the verified badge, and you can upgrade the listing to be featured on localhosted. Proud to be listed? Put our badge on your site.
Want more people to find MetaGPT?Promote it
Something wrong or outdated on this page?
33.7KUpdated 4 months ago
Windows · Docker#Code execution#Human approval#Multi-agent workflows
GPT Pilot is a self-hosted AI coding assistant for developers who want an agent to build applications under their supervision. The project is no longer maintained. It uses the FSL-1.1-MIT source-available license, with restrictions on competing uses. It runs locally as a Python CLI and is the core technology behind the Pythagora VS Code extension.
55.1KUpdated 2 years agoMIT
Windows · Docker#Code execution#Multimodal input
42Updated 12 hours ago
macOS · Windows · Linux · Docker#Agent Skills#Code execution#Distributed execution
110.7KUpdated 3 hours agoMIT
macOS · Windows · Linux · Docker#Agent Skills#Code execution#MCP
34.4KUpdated 2 months agoApache-2.0
macOS · Windows · Linux · Docker · Web#Batch processing#Human approval#MCP
15.7KUpdated 12 months agoMIT
Windows · Docker#Code execution#Git integration#Human approval
gpt-engineer is a locally run coding assistant for developers who want to experiment with AI code generation and build their own agents. It can create software from plain-language descriptions or make requested changes to an existing codebase. The project is archived and no longer maintained. Its Python code is open source under the MIT license.
Code Buddy is an AI coding assistant for developers who want a terminal agent running on their own machine, with a choice of local or cloud models. It reads repositories, edits code and runs commands on Linux, macOS and Windows. Ollama keeps model inference local without an API key or account; cloud providers send model requests off the machine. Routing includes automatic provider failover.
Pi is a terminal-based coding assistant for developers who want to shape the agent around their own workflow. Its core stays small, while extensions can change its tools, commands and terminal interface. It's open source under the MIT license.
ChatDev is a self-hosted platform for building teams of AI agents through a visual workflow editor. It's for people who want agents to collaborate on research, data analysis or software projects without writing the orchestration code themselves. You define each agent's role and how information passes between them.
Plandex is an open source, terminal-based AI coding agent for developers working on changes that span many files. Its review sandbox holds generated edits apart from your project files so you can inspect the accumulated changes before applying them. You can run its server locally through Docker or host it on your own server. The code uses the MIT license.