---
title: Use containers for generative AI development
linkTitle: Develop your app
weight: 20
keywords: python, local, development, generative ai, genai, llm, neo4j, ollama, langchain, openai
description: Learn how to develop your generative AI (GenAI) application locally.
aliases:
  - /guides/use-case/genai-pdf-bot/develop/
---

## Prerequisites

Complete [Containerize a generative AI application](containerize.md).

## Overview

In this section, you'll learn how to set up a development environment to access all the services that your generative AI (GenAI) application needs. This includes:

- Adding a local database
- Adding a local or remote LLM service

> [!NOTE]
>
> You can see more samples of containerized GenAI applications in the [GenAI Stack](https://github.com/docker/genai-stack) demo applications.

## Add a local database

You can use containers to set up local services, like a database. In this section, you'll update the `compose.yaml` file to define a database service. In addition, you'll specify an environment variables file to load the database connection information rather than manually entering the information every time.

To run the database service:

1. In the cloned repository's directory, rename `env.example` file to `.env`.
   This file contains the environment variables that the containers will use.
2. In the cloned repository's directory, open the `compose.yaml` file in an IDE or text editor.
3. In the `compose.yaml` file, add the following:

   - Add instructions to run a Neo4j database
   - Specify the environment file under the server service in order to pass in the environment variables for the connection

   The following is the updated `compose.yaml` file. All comments have been removed.

   ```yaml{hl_lines=["7-23"]}
   services:
     server:
       build:
         context: .
       ports:
         - 8000:8000
       env_file:
         - .env
       depends_on:
         database:
           condition: service_healthy
     database:
       image: neo4j:5.11
       ports:
         - "7474:7474"
         - "7687:7687"
       environment:
         - NEO4J_AUTH=${NEO4J_USERNAME}/${NEO4J_PASSWORD}
       healthcheck:
         test: ["CMD-SHELL", "wget --no-verbose --tries=1 --spider localhost:7474 || exit 1"]
         interval: 5s
         timeout: 3s
         retries: 5
   ```

   > [!NOTE]
   >
   > To learn more about Neo4j, see the [Neo4j Official Docker Image](https://hub.docker.com/_/neo4j).

4. Run the application. Inside the `docker-genai-sample` directory,
   run the following command in a terminal.

   ```console
   $ docker compose up --build
   ```

5. Access the application. Open a browser and view the application at [http://localhost:8000](http://localhost:8000). You should see a simple Streamlit application. Note that asking questions to a PDF will cause the application to fail because the LLM service specified in the `.env` file isn't running yet.

6. Stop the application. In the terminal, press `ctrl`+`c` to stop the application.

## Add a local or remote LLM service

The sample application supports both [Ollama](https://ollama.ai/) and [OpenAI](https://openai.com/). This guide provides instructions for the following scenarios:

- Run Ollama in a container
- Run Ollama outside of a container
- Use OpenAI

While all platforms can use any of the previous scenarios, the performance and
GPU support may vary. You can use the following guidelines to help you choose the appropriate option:

- Run Ollama in a container if you're on Linux, and using a native installation of the Docker Engine, or Windows 10/11, and using Docker Desktop, you
  have a CUDA-supported GPU, and your system has at least 8 GB of RAM.
- Run Ollama outside of a container if you're on an Apple silicon Mac.
- Use OpenAI if the previous two scenarios don't apply to you.

Choose one of the following options for your LLM service.

{{< tabs >}}
{{< tab name="Run Ollama in a container" >}}

When running Ollama in a container, you should have a CUDA-supported GPU. While you can run Ollama in a container without a supported GPU, the performance may not be acceptable. Only Linux and Windows 11 support GPU access to containers.