DIAL Actions Demo Application

To run the DIAL application locally, please follow the instructions below.

Prerequisites

• Poetry
• Docker
• Docker Compose

Installation

Common Steps

1. Clone the repository

   git clone

2. Install dependencies

   poetry install --no-root --directory=app

3. Create .env.chat file

   DIAL_API_KEY='dial_api_key' # should be the same as in the config.json
   NEXTAUTH_SECRET='secret'
   DEFAULT_MODEL=gpt-35-turbo

4. Create .env.app file with the actual values according to this documentation

5. Create .env.core file if you want to use models

   AZURE_OPENAI_ENDPOINT='AZURE_OPENAI_ENDPOINT'
   AZURE_OPENAI_API_KEY='AZURE_OPENAI_API_KEY'
   
   AZURE_DEPLOYMENT_OPENAI_35=gpt-35-turbo-16k
   AZURE_DEPLOYMENT_OPENAI_4=gpt-4o
   AZURE_DEPLOYMENT_EMBEDDING=text-embedding-ada-002
   AZURE_OPENAI_API_VERSION=2024-02-01

   Important: These variables are used in /dial/config/config-template.json. If you want to use other variables, update the /tools/update-dial-config.sh script accordingly.

Running the Application

Without Colima

1. Run the application

   docker-compose up

With Colima

1. Start Colima

   colima start --arch x86_64 --memory 8 --cpu 4

2. Run init.sh script

   ./init.sh

3. Run the application

   docker-compose -f docker-compose-colima.yml up

4. Wait for the application to start (This may take some time)

   dial-app-1        | INFO:     Started server process [1]
   dial-app-1        | INFO:     Waiting for application startup.
   dial-app-1        | INFO:     Application startup complete.
   dial-app-1        | INFO:     Uvicorn running on http://0.0.0.0:5002 (Press CTRL+C to quit)

5. Open the browser and go to http://localhost:3000/

Usage

1. Open the browser and go to http://localhost:3000/
2. Select the application to use and start the conversation.
3. To customize the response, find the relevant code in app/application/dial_application.py and update it according to your needs.