Quick Start
Get Hanflow running locally in under five minutes using Docker.
Prerequisites
- Docker 24+ and Docker Compose v2
- A model API key (OpenAI, GLM, DeepSeek, Anthropic, or a local Ollama/vLLM server)
1. Clone
git clone https://github.com/hanflow/hanflow
cd hanflow
2. Configure
Copy the example config and environment files, then put your model keys in .env.
cp deploy/docker/hanflow.yaml.example hanflow.yaml
cp deploy/docker/.env.example .env
The example config defines a strong model and a fast model, with the strong
model as the routing default and the fast model as fallback.
models:
strong: { provider: openai, model: gpt-4o, api_key: "${OPENAI_KEY}" }
fast: { provider: glm, model: glm-4-flash, api_key: "${GLM_KEY}" }
routing:
default: strong
fallback_chain: [strong, fast]
3. Run
Start the full stack (engine, web studio, postgres, redis) with one command.
cd deploy/docker
docker compose up -d
Open the services:
- Web Studio: http://localhost:3000
- API and interactive docs: http://localhost:8000/docs
Verify
Create a workflow in Web Studio, or run one from the CLI:
hanflow validate workflows/untitled.yaml
hanflow run workflows/untitled.yaml
Use hanflow doctor to check your environment and config health.