Skip to main content

🚀 Deploying flux0 to Koyeb

Koyeb is a great platform for hosting Flux0 in the cloud with minimal setup. it is a great choice for hosting production-ready flux0 agents with zero infrastructure overhead. This guide walks you through deploying Flux0 via GitHub as a Koyeb service.

🌟 Why Koyeb?​

Koyeb is a great match for flux0 agents. Here's why:

  • Git-powered deploys Push your agents to GitHub and Koyeb deploys it automatically — no CI/CD needed.

  • Built-in secret management Store API keys like OPENAI_API_KEY securely as environment variables.

  • Zero infrastructure No servers, no containers — just your agents code running in the cloud.

  • Global performance Agents run behind a CDN and auto-scale with traffic, out of the box.

  • HTTPS & custom domains Every agent is live on HTTPS by default, with custom domain support.

Koyeb handles the deploys, flux0 handles the rest — from multi-agent logic to session state and LLM calls. You focus on agent behavior, hit git push, and it’s live.

🧾 What You’ll Need​

  • A GitHub repository with your flux0 app
  • A Koyeb account
  • Python 3.13+ compatibility
  • LLM (e.g, OpenAI) API key

🚀 Deployment Steps​

🔐 Step1: Create a Secret​

To securely store sensitive values as the API key:

  1. Go to your Koyeb Dashboard.
  2. Navigate to Secrets in the sidebar.
  3. Click "Create Secret".
  4. Name your secret (e.g., openai-api-key) and paste the value.
  5. Save it.

Create a Secret

🧠 Step2: Agent Logic​

Your agent logic should live inside a folder under modules, for example modules/my_agent.

Step 3: Push to GitHub​

Make sure your flux0 app is in a public or private GitHub repository.

tip

You can fork https://github.com/flux0-ai/flux0-docker-example with your agent code or use this repo directly in Koyeb for demonstration.

Step4: Create a Koyeb App (Service)​

  1. Go to app.koyeb.com
  2. Click Create Service
  3. Choose:

Create Service

Choose a repo​

Connect your private repo or choose a public one and click import.

Choose repo

Configure service​

Select a CPU of your choice and click Next (CPU Eco should be sufficient for demonstration purposes)

Builder​

Define Builder to Dockerfile and leave all settings as default.

Environment variables​
KeyExample Value
OPENAI_API_KEY{{ secret.openai-api-key }}
FLUX0_MODULESmy_agent
Exposed Ports​

Change port to 8080

✅ Step5: Verify Your Deployment​

  • Click Deploy
  • Wait for the service to be deployed.
  • Click the Public URL, you should see the chat interaface.

🧾 Step 6: Register the Agent​

Once your runner is defined and your module is registered, you can create the agent entry in the database:

  1. Cick manage your service.
  2. Click Console and press Enter
flux0 agents create --name "Translation Agent" --type langchain_simple

💬 Step 7: Talk to Your Agent​

Start chatting with your agent at the service URL.

📚 Related​