AdGen AI

AI-Powered Video & Image Generation

📐 Try Frame Grid View

AI Video Generator

PNG, JPG, GIF up to 10MB

Direct Image Generation

Try It Now

Live Demo

💡 Tips for better results:

  • Be specific and descriptive in your prompts
  • Include details about style, lighting, and mood
  • Watch as your image generates in real-time!

API Endpoint

POST /api/generate-image

Powered by: OpenAI Responses API with GPT-5 + image_generation tool

Parameters

  • prompt:(required) Description of the image to generate or edit
  • image / image[]:(optional) Single or multiple image files for editing
  • previous_response_id:(optional) For multi-turn iterative refinement
  • quality:(optional) low, medium, high, or auto (default: auto)
  • background:(optional) transparent, opaque, or auto (default: auto)
  • input_fidelity:(optional) low or high (default: low)

Advanced Generation

Generate stunning images using GPT-5 with superior quality and instruction following

Multi-Image Editing

Use multiple reference images with high fidelity preservation of details and faces

Multi-turn Refinement

Iteratively refine images across multiple requests for perfect results

Quick Start Examples

1. Generate an image

curl -X POST http://localhost:3000/api/generate-image \
  -F "prompt=A futuristic city at sunset with flying cars"

2. Generate with transparent background

curl -X POST http://localhost:3000/api/generate-image \
  -F "prompt=A cute robot character" \
  -F "quality=high" \
  -F "background=transparent"

3. Edit with multiple images (high fidelity)

curl -X POST http://localhost:3000/api/generate-image \
  -F "prompt=Create a gift basket with these items" \
  -F "image[0]=@product1.png" \
  -F "image[1]=@product2.png" \
  -F "input_fidelity=high"

4. Multi-turn refinement (saves response_id)

# First request
RESP=$(curl -X POST http://localhost:3000/api/generate-image \
  -F "prompt=A cat hugging an otter")
ID=$(echo $RESP | jq -r '.response_id')

# Refine the image
curl -X POST http://localhost:3000/api/generate-image \
  -F "prompt=Make it photorealistic" \
  -F "previous_response_id=$ID"

⚠️ Setup Required

Make sure to add your OpenAI API key to .env.local before using the API. See the README.md for detailed instructions.