#5050th of 80 in Agent platforms
chatdev
Visual builder for multi-agent workflows defined in YAML
- Stars
- 34.7k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Mar 2026
- Language
- Python
32,436 on 20 Apr 202634,725 on 9 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
ChatDev 2.0 (DevAll) is a FastAPI backend with a Vue 3 web console for designing and running multi-agent workflows on a drag-and-drop canvas. Workflows are YAML files, runnable from the Launch tab or through a Python SDK (`chatdev` on PyPI). Bundled templates cover data visualization, Blender 3D generation, game development, deep research and teaching videos. The original ChatDev 1.0 software-company simulation lives on the `chatdev1.0` branch.
Who it is for: Engineers building multi-agent workflows without writing orchestration code
Strengths
- Visual canvas plus YAML configs, checkable with `make validate-yamls`
- Ships workflows for data visualization, deep research, game dev, Blender 3D and teaching video
- Python SDK `run_workflow` supports attachments and variable overrides
- Custom Python tools go in `functions/`; human-in-the-loop feedback at launch
Weaknesses
- Needs Python 3.12+, Node.js 18+ and uv; backend and frontend run as two services
- 3D workflows also require Blender and blender-mcp installed separately
- Needs an external LLM provider key and base URL; no bundled model runtime
- ChatDev 1.0 software-company mode is legacy, moved to another branch
What it needs
- no GPU
- Docker + Compose
- Needs Python 3.12+, Node.js 18+, uv, LLM provider API (API_KEY, BASE_URL), Blender and blender-mcp (3D workflows only), manim (teach video workflow only)
- Models: LLM providers via API_KEY and BASE_URL
- port 5173
Quick answers
Is chatdev open source?
Yes. Its code is on GitHub under the Apache-2.0 license.
Can you run chatdev with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does chatdev need a GPU?
No, it runs without one.
Also in Agent platforms
See all 80| Rank | Project | Score |
|---|---|---|
| 1 | AutoGPTRole-based AI agents coordinated by one lead agent, plus a visual workflow builder | 95 out of 100 |
| 2 | daily_stock_analysisLLM-driven daily analysis of watchlist stocks across six markets, pushed to chat apps | 95 out of 100 |
| 3 | LangflowVisual builder for AI agents and workflows, exposed as API or MCP server | 94 out of 100 |
| #4 | deer-flowAgent harness running sub-agents, memory and sandboxes with pluggable skills | 93 out of 100 |
| #5 | DifyVisual builder for LLM workflows, RAG pipelines and agents, with APIs | 92 out of 100 |