Skip to main content
Version: dev

Quick Install

The fastest way to get DB-GPT running. The installer script prepares a local DB-GPT workspace, generates a provider profile, and gives you a ready-to-run webserver command.

Use the installer script if you want the shortest path from zero to a working DB-GPT web UI.

curl -fsSL https://raw.githubusercontent.com/eosphoros-ai/DB-GPT/main/scripts/install/install.sh | bash

System requirements​

This quick-install flow is designed for:

  • macOS or Linux
  • a shell environment that can run bash
  • network access to download dependencies
  • an API key if you plan to use a hosted model provider immediately
Best fit

Choose this path if you want to try DB-GPT quickly without managing the repository structure yourself.

Install with a provider profile​

If you already know which provider you want, pass the profile and API key directly during installation.

OpenAI-compatible profile​

curl -fsSL https://raw.githubusercontent.com/eosphoros-ai/DB-GPT/main/scripts/install/install.sh   | OPENAI_API_KEY=sk-xxx bash -s -- --profile openai

Kimi 2.5 via Moonshot API​

curl -fsSL https://raw.githubusercontent.com/eosphoros-ai/DB-GPT/main/scripts/install/install.sh   | MOONSHOT_API_KEY=sk-xxx bash -s -- --profile kimi

MiniMax via an OpenAI-compatible API​

curl -fsSL https://raw.githubusercontent.com/eosphoros-ai/DB-GPT/main/scripts/install/install.sh   | MINIMAX_API_KEY=sk-xxx bash -s -- --profile minimax

Reuse an existing local checkout​

Already have a local DB-GPT repository? Reuse it instead of cloning into ~/.dbgpt/DB-GPT.

Reuse local repo with OpenAI​

OPENAI_API_KEY=sk-xxx   bash scripts/install/install.sh --profile openai --repo-dir "$(pwd)" --yes

Reuse local repo with Kimi​

MOONSHOT_API_KEY=sk-xxx   bash scripts/install/install.sh --profile kimi --repo-dir "$(pwd)" --yes

Reuse local repo with MiniMax​

MINIMAX_API_KEY=sk-xxx   bash scripts/install/install.sh --profile minimax --repo-dir "$(pwd)" --yes

What the installer prepares​

The installer script sets up the common runtime layout for you:

  • a DB-GPT checkout under ~/.dbgpt/DB-GPT unless --repo-dir is used
  • generated provider configs under ~/.dbgpt/configs/
  • the DB-GPT home directory under ~/.dbgpt/
  • a ready-to-run webserver command using the generated profile

Start DB-GPT after installation​

After installation completes, start the webserver with the generated profile config:

cd ~/.dbgpt/DB-GPT && uv run dbgpt start webserver --profile <profile>

Then open http://localhost:5670.

Verify the install​

Your install is working if:

  • the webserver starts without configuration errors
  • the Web UI opens at http://localhost:5670
  • you can start a chat in the browser

Review the script first​

If you prefer to inspect the installer before running it:

curl -fsSL https://raw.githubusercontent.com/eosphoros-ai/DB-GPT/main/scripts/install/install.sh -o install.sh
less install.sh
bash install.sh --profile openai

Alternative install methods​

If the installer script is not the right fit for your environment:

  • use CLI Install for a PyPI-based install with the dbgpt command
  • use Source Install for development, debugging, and customization

Troubleshooting​

The installer script does not match my shell or platform​

Use CLI Install or Source Install instead.

I want more control over dependencies and configuration​

Use Source Install. It exposes the full repository layout and uv sync workflow.

The install completed, but DB-GPT does not start cleanly​

Check the generated config under ~/.dbgpt/configs/, then see Installation Issues.