Spring AI Powered API for Chatbot, Image Generation, and Recipe Generation

This project explores the capabilities of Spring AI to build intelligent applications leveraging OpenAI models. It provides APIs for:

• Chatbot: Engaging in conversational interactions using GPT-4.0 or GPT-3.5.
• Image Generation: Creating images based on text prompts using DALL-E 2 or DALL-E 3.
• Recipe Generation: Generating recipes based on provided ingredients, dietary restrictions, and cuisine preferences.

Technologies Used

• Java: The primary programming language.
• Spring Boot: Framework for building the application.
• Spring AI: Abstraction layer for interacting with various AI models.
• OpenAI API: Accessing GPT and DALL-E models.
• Maven: Build automation tool.
• (Optional) Lombok: For reducing boilerplate code.
• (Optional) Spring Web: For building RESTful APIs.

Prerequisites

• Java Development Kit (JDK): Ensure you have a compatible JDK installed.
• Maven: Install Maven if you don't have it already.
• OpenAI API Key: You will need an API key from OpenAI to access their models. You can obtain one from the OpenAI Platform.

Setup and Installation

1. Clone the Repository:

   git clone <(https://github.com/keshavarun20/spring-ai-multitool/)>
   cd <YOUR_PROJECT_DIRECTORY>

2. Configure OpenAI API Key:

   • Create an application.properties or application.yml file in the src/main/resources directory.
   • Add your OpenAI API key:

     spring.ai.openai.api-key=<YOUR_OPENAI_API_KEY>

     or

     spring:
       ai:
         openai:
           api-key: <YOUR_OPENAI_API_KEY>

3. Build the Project:

   mvn clean install

Running the Application

1. Navigate to the Project Directory:

   cd <YOUR_PROJECT_DIRECTORY>

2. Run the Spring Boot Application:

   mvn spring-boot:run

   The application will typically start on http://localhost:8080.

API Endpoints (as seen in Postman/Browser)

When making requests to these endpoints, you will construct the request URL with parameters. The successful responses (typically with a 200 OK status code) will contain the generated content in the body as JSON.

Chatbot

• Endpoint: /api/ask-ai (GET)
• Request Parameters (in URL):
  • prompt: The message you want to send to the chatbot (e.g., prompt=Tell me a joke).
• Example Request URL: http://localhost:8080/api/ask-ai?prompt=Tell%20me%20a%20joke
• Successful Response (200 OK) Body (JSON):

  {
    "response": "Why don't scientists trust atoms? Because they make up everything!"
  }

  Postman/Browser Observation: Construct the URL with the prompt parameter. The JSON response containing the response will be in the "Body" tab of the Postman response or rendered by the browser.

Image Generation

• Endpoint: /api/generate-image (GET)
• Request Parameters (in URL):
  • prompt: The description of the image you want to generate (e.g., prompt=A futuristic cityscape).
• Example Request URL: http://localhost:8080/api/generate-image?prompt=A%20futuristic%20cityscape
• Successful Response (200 OK) Body (JSON):

  {
    "imageUrl": "URL_OF_THE_GENERATED_IMAGE"
  }

  Postman/Browser Observation: Build the URL with prompt and optionally model parameters. The imageUrl will be in the JSON response body.

Recipe Generation

• Endpoint: /api/generate-recipe (GET)
• Request Parameters (in URL):
  • ingredients: A comma-separated list of ingredients (e.g., ingredients=chicken,broccoli,rice).
  • dietaryRestrictions (Optional): A comma-separated list of dietary restrictions (e.g., dietaryRestrictions=gluten-free,vegetarian).
  • cuisine (Optional): The desired cuisine (e.g., cuisine=Italian).
• Example Request URL: http://localhost:8080/api/generate-recipe?ingredients=chicken,broccoli,rice&dietaryRestrictions=gluten-free&cuisine=Italian
• Successful Response (200 OK) Body (JSON):

  {
    "recipeName": "Chicken and Broccoli Risotto",
    "ingredients": ["chicken", "broccoli", "arborio rice", ...],
    "instructions": ["...", "...", "..."]
  }

  Postman/Browser Observation: Construct the URL with the ingredients, dietaryRestrictions, and cuisine parameters. The recipe details will be in the JSON response body.