Gemini Flash API

A Node.js Express API that integrates with Google's Gemini AI to generate content from text, images, documents, and audio files.

Features

• Text Generation: Generate AI responses from text prompts
• Image Analysis: Upload images and get AI-generated descriptions or analysis
• Document Processing: Upload documents and get summaries or analysis
• Audio Transcription: Upload audio files and get transcriptions

Prerequisites

• Node.js (v14 or higher)
• Google AI API Key (Gemini)

Installation

1. Clone the repository:

git clone https://github.com/hanifalkauni/gemini-ai-api-project.git
cd gemini-flash-api

2. Install dependencies:

npm install

3. Set up environment variables:

cp .env.example .env

4. Add your Google AI API key to the .env file:

GEMINI_API_KEY=your_google_ai_api_key_here
PORT=3000

Usage

Start the server:

node index.js

The server will run on http://localhost:3000 (or the port specified in your .env file).

API Endpoints

1. Generate Text

POST /generate-text

Generate AI content from a text prompt.

Request Body:

{
  "prompt": "Your text prompt here"
}

Response:

{
  "result": "AI generated response"
}

2. Generate from Image

POST /generate-from-image

Upload an image and get AI analysis or description.

Request:

• Content-Type: multipart/form-data
• Fields:
  • image: Image file (required)
  • prompt: Text prompt for image analysis (optional)

Response:

{
  "result": "AI analysis of the image"
}

3. Generate from Document

POST /generate-from-document

Upload a document and get AI summary or analysis.

Request:

• Content-Type: multipart/form-data
• Fields:
  • document: Document file (required)
  • prompt: Text prompt for document analysis (optional)

Response:

{
  "result": "AI summary or analysis of the document"
}

4. Generate from Audio

POST /generate-from-audio

Upload an audio file and get transcription.

Request:

• Content-Type: multipart/form-data
• Fields:
  • audio: Audio file (required)
  • prompt: Text prompt for audio processing (optional)

Response:

{
  "result": "Audio transcription"
}

Example Usage

Using curl

Text Generation:

curl -X POST http://localhost:3000/generate-text \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Explain quantum computing in simple terms"}'

Image Analysis:

curl -X POST http://localhost:3000/generate-from-image \
  -F "image=@path/to/your/image.jpg" \
  -F "prompt=Describe this image"

Document Processing:

curl -X POST http://localhost:3000/generate-from-document \
  -F "document=@path/to/your/document.pdf" \
  -F "prompt=Summarize this document"

Audio Transcription:

curl -X POST http://localhost:3000/generate-from-audio \
  -F "audio=@path/to/your/audio.mp3" \
  -F "prompt=Transcribe this audio"

Dependencies

• express: Web framework for Node.js
• multer: Middleware for handling multipart/form-data (file uploads)
• @google/genai: Google AI SDK for Gemini integration
• dotenv: Environment variable management

File Structure

gemini-flash-api/
├── uploads/          # Temporary file upload directory
├── .env             # Environment variables (not in repo)
├── .env.example     # Environment variables template
├── .gitignore       # Git ignore rules
├── index.js         # Main application file
├── package.json     # Project dependencies and scripts
├── package-lock.json # Dependency lock file
└── README.md        # This file

Error Handling

The API includes error handling for:

• Invalid requests
• File upload errors
• Google AI API errors
• Server errors

All errors return a JSON response with an error message:

{
  "error": "Error description"
}

Security Notes

• Uploaded files are automatically deleted after processing
• Keep your Google AI API key secure and never commit it to version control
• The uploads directory is created automatically if it doesn't exist

License

ISC