Skip to main content

Starter Tutorial

Create a Story Generator application​

In this tutorial we will create a simple story generator application. The application will generate a story based on a theme provided by the user. The application will use ChatGPT to generate a story and Dall-E to generate an image for the story.

Start LLMStack​

Clone the repository

git clone https://github.com/trypromptly/LLMStack.git

Compile the frontend code

cd LLMStack/client
npm run build

Build and start the docker image

docker-compose -f docker-compose.dev.yml -d --build --env-file .env.dev up

In you browser visit http://localhost:9000 and login with the default admin account.

[TODO: Add screenshot of login page]

Update API Keys​

The application will use OpenAI's ChatGPT and OpenAI's Dall-E to generate the story and image respectively. Both these models require an API key to access the API. To update the API key, visit the settings page and add your OpenAI key.

[TODO: Add screenshot of settings page]

Create a new application​

To create a new application click on Web App under Create a new App from scratch section on the home page. This will open the application builder UI. [TODO: Add steps]

Preview the application​

Upon completing the above steps, hit save. Now you can preview your app by clicking the Preview item in the left hand side menu. This will render your web app in the existing browser tab. Now enter the theme for the story and click on the Submit button. ✨Voila!✨ You have generated your first story.

Publish the application​

You can continue iterating on your prompt to improve the story generation. Once you are happy with the results, you can publish the application by clicking on the Publish button. This will make the application available to others to use.