ChatDev
byOpenBMBlisted Jun 3, 2026
ChatDev is a research project from OpenBMB (Tsinghua University) that demonstrates a virtual software company where multiple LLM agents with distinct roles — CEO, CTO, programmer, reviewer, tester — collaborate to build software from a single natural language requirement. It explores how software engineering processes can be fully automated through role-playing agents and structured communication protocols. ChatDev remains an influential research artifact and educational tool for understanding multi-agent systems.
Freshness note: AI-agent behavior can change after model releases, client updates, pricing changes, or new default settings. Use the aggregate score as a historical baseline and check recent reviews plus model/runtime context before deciding.
- Pricing
- Open-source research project; you pay for your own OpenAI API keys
- Model backbone
- GPT-4, GPT-3.5-turbo
- Setup complexity
- Moderate setup
- Last reviewed
- Awaiting reviews
- Recent signal
- Awaiting reviews
What reviewers consistently say.
No reviewer patterns for ChatDev yet.
No consensus yet
Once people review ChatDev, Ruling will summarize recurring strengths and limitations here.
Write the first reviewNo community reviews yet.
Used ChatDev? Your field report can seed the first Ruling signal.
Be the first to review ChatDev
Share where it worked, where it failed, and whether it deserves a spot in a real team workflow.
Write a reviewSimilar agents.
Frameworks & Indie agents you might also consider, sorted by aggregate score.
AutoGPT was the pioneering autonomous agent project that sparked the 2023 AI agent explosion, demonstrating that LLMs could self-direct toward goals using a loop of thought, action, and memory. With over 170,000 GitHub stars, it remains one of the most starred AI repos ever and introduced concepts like agent memory and tool use that are now standard in the field. Its AutoGPT Platform continues development as both an open-source framework and hosted cloud agent service.
Open-source orchestration runtime for agentic workloads. Defines tasks, workspaces, network policies, and model configuration, with sandboxed execution, resource limits, and suspend/resume support. Runs on Agent Substrate and provides Kubernetes deployment manifests for operating agent fleets.
BabyAGI is the lightweight task-driven autonomous agent that inspired a generation of agent developers — using a simple loop of task creation, prioritization, and execution to work toward user-defined goals. Despite its simplicity (the original was 140 lines of Python), it demonstrated that effective goal-directed behavior could emerge from basic LLM loops. The project remains influential as a conceptual reference point, though most production use cases have moved to more mature frameworks.