BuildingAI is an open-source enterprise-grade platform that lets developers configure and deploy AI-powered enterprise apps via Docker, with features like AI conversations, agents, knowledge bases, RAG, and billing. It provides a visual configuration interface to build AI skills without code.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
BuildingAI is an enterprise-grade open-source intelligent agent platform designed for AI developers, AI entrepreneurs, and forward-thinking organizations. Through a visual configuration interface (Do It Yourself), you can build native enterprise AI applications without code. The platform offers native capabilities such as intelligent agents, MCP, RAG pipelines, knowledge bases, large-model aggregation, and context engineering, along with user registration, membership subscriptions, compute billing, and other business operations.
How it works
The project provides a Docker-based deployment approach. Start by copying environment variables, then run docker compose up -d. The README describes docker deployment steps and notes that the build progress can be monitored in Node.js container logs until an accessible URL appears, indicating the project started.
Key capabilities include AI Conversations (LLMs with multimodal support), AI Agents (with memory, goals, and tool usage), Knowledge Bases with vector search and RAG, MCP Integration via SSE and Streamable HTTP, a unified API for models, an Extension mechanism, and built-in billing/Payments.
Getting started
# Enter the project directory (replace with your directory name)
dc buildingai
# Copy and customize the environment variables
# In production, update the APP_DOMAIN value in the .env file to your own domain.
cp .env.example .env
# Start the application with Docker
docker compose up -d
After startup, open http://localhost:4090/install to complete initial setup. For other deployment methods, see Deployment Guide.
Recent releases
Latest five releases show updates labeled 26.1.1 (2026-05-15), 26.1.0 (2026-04-30), 26.0.3 (2026-04-10), 26.0.2 (2026-04-10), 26.0.1 (2026-04-02). Each release contains a Docker Deployment section with commands to pull the latest code and restart the Node.js container, e.g.,
Pull the latest code via Git: `git reset --hard HEAD && git pull`
Restart the Node.js container under `buildingai` in Docker:
`docker restart buildingai-nod`
Traction
1846 stars, 447 forks, 74 open issues. Language: TypeScript. License: Apache-2.0.
Behind the repo
The project is hosted at GitHub under BidingCC/BuildingAI with topics ai, coze, dify, fastgpt, llm, rag.
Caveats
License: Apache License 2.0. Created 2025-03-14. Last push 2026-06-08.