Getting Started with SuperAGI: Docker Setup and Your First Autonomous Agent

Prerequisites

Before you begin, ensure you have the following installed:

Docker and Docker Compose (v2.20+)
At least 4 GB RAM and 2 CPU cores
An OpenAI API key (or another LLM provider key)
Git for cloning the repository

Step 1: Clone the Repository

Open your terminal and clone the SuperAGI repository:

git clone https://github.com/TransformerOptimus/SuperAGI.git
cd SuperAGI

Step 2: Configure Environment Variables

Create a copy of the configuration template and add your API keys:

cp config_template.yaml config.yaml

Edit config.yaml and set your LLM provider details:

llm:
  provider: openai
  api_key: YOUR_OPENAI_API_KEY
  model: gpt-4
  
vector_db:
  provider: pinecone
  api_key: YOUR_PINECONE_API_KEY
  environment: your-environment

Step 3: Launch with Docker Compose

Build and start SuperAGI with Docker Compose. The standard setup uses PostgreSQL, Redis, and other supporting services automatically:

docker compose -f docker-compose.yaml up --build -d

If you have a GPU and want to use local LLMs, use the GPU variant instead:

docker compose -f docker-compose-gpu.yml up --build -d

The initial build may take a few minutes as it compiles the Python dependencies and sets up the database schema.

Step 4: Access the Web Interface

Once the containers are running, open your browser and navigate to:

http://localhost:3000

You will see the SuperAGI login screen. Create an account or log in using your credentials. The dashboard presents your agents, toolkits, and performance metrics at a glance.

Step 5: Create Your First Agent

From the dashboard, click Create Agent. Give your agent a name and a clear goal — for example, "Research the latest AI trends and summarize them." Configure the following:

Model: Choose GPT-4, Claude, or your local LLM
Tools: Enable web search, web scraping, and file manager
Memory: Select your vector database for long-term storage
Workflow: Use ReAct for multi-step reasoning

Once configured, click Run. The Action Console will show the agent thinking through each step, selecting tools, and producing results in real time.

Why Run SuperAGI on Your Own Hardware?

Self-hosting SuperAGI gives you full control over your data, agent configurations, and cost. Unlike managed agent services, self-hosting means:

Your agent data stays private — no third-party servers involved
Connect any LLM — OpenAI, Anthropic, Ollama, llama.cpp, or custom models
No rate limits — scale agents as needed
Full customization of toolkits, workflows, and agent behaviors

Troubleshooting

If you encounter issues:

Port conflict: Change the host port in docker-compose.yaml (e.g., 3001:3000)
Database errors: Ensure PostgreSQL container is healthy (docker ps)
API key errors: Double-check your config.yaml for typos or missing keys
Build fails: Try docker compose down --volumes && docker compose up --build

For further help, check the official documentation or join the SuperAGI Discord community.