# GPT Image 1.5 - Core.Today AI API

> OpenAI's latest image generation model with better instruction following and adherence to prompts, including sharp text rendering and detailed editing.

- **Provider**: OpenAI
- **Model ID**: openai/gpt-image-1.5
- **Category**: Image Generation
- **Credits**: 310 per image at high/auto quality; 31 at low, 120 at medium
- **Speed**: Medium
- **Quality**: Ultra

## Features
- Four quality tiers (low/medium/high/auto) with proportional pricing
- Strong instruction following and prompt adherence
- Sharp in-image text rendering
- Image editing via input_images with adjustable input_fidelity
- Transparent or opaque background control

## Use Cases
- Marketing graphics with legible embedded text
- Product mockups with transparent backgrounds
- Low-cost draft iteration using the low/medium quality tiers
- High-fidelity image editing that preserves facial features

## API Endpoint

Base URL: https://api.core.today/v1
Create Prediction: POST /predictions
Get Status: GET /predictions/{job_id}

## Authentication
Header: X-API-Key: YOUR_API_KEY

## Input Parameters

### Required
- **prompt**: string - A text description of the desired image

### Optional
- **openai_api_key**: string - Your OpenAI API key (optional - uses proxy if not provided)
- **aspect_ratio**: string (default: 1:1) - The aspect ratio of the generated image
  Options: 1:1, 3:2, 2:3
- **input_fidelity**: string (default: low) - How much effort the model exerts to match style and facial features of input images
  Options: low, high
- **input_images**: array - A list of images to use as input for editing or composing
- **number_of_images**: integer (default: 1) - Number of images to generate. Restricted to 1 in this gateway for accurate per-image billing
- **quality**: string (default: auto) - The quality of the generated image. Determines the credit tier (low/medium/high/auto)
  Options: low, medium, high, auto
- **background**: string (default: auto) - Set whether the background is transparent, opaque, or chosen automatically
  Options: auto, transparent, opaque
- **output_compression**: integer (default: 90) - Compression level (0-100%)
- **output_format**: string (default: webp) - Output image format
  Options: png, jpeg, webp
- **moderation**: string (default: auto) - Content moderation level
  Options: auto, low
- **user_id**: string - An optional unique identifier representing your end-user. This helps OpenAI monitor and detect abuse.

## Examples

### High-Quality Photorealistic Scene (Default)
A detailed photorealistic scene at high quality
```json
{
  "model": "openai/gpt-image-1.5",
  "input": {
    "prompt": "make a scene in chelsea, london in the 1970s, photorealistic, everything in focus, with tons of people, and a bus with an advertisement for \"ImageGen 1.5 on Replicate\"",
    "quality": "high",
    "aspect_ratio": "1:1"
  }
}
```

### Low-Cost Draft
Fast, cheap iteration using the low quality tier
```json
{
  "model": "openai/gpt-image-1.5",
  "input": {
    "prompt": "a minimalist logo concept for a coffee brand, flat design",
    "quality": "low"
  }
}
```

## Response Format
```json
{
  "job_id": "abc123",
  "status": "pending | processing | completed | failed",
  "result": "URL or data (when completed)"
}
```

## Usage Flow
1. POST /predictions with model and input -> receive job_id
2. GET /predictions/{job_id} -> poll until status is completed or failed
3. Result contains output URL(s)

## Tips
- Set quality explicitly to low or medium while iterating — auto/high are billed at the top rate (510 credits)
- number_of_images is capped at 1 in this gateway; call the API multiple times for multiple images
- Use input_fidelity=high when editing images that must preserve facial features precisely
- background=transparent is useful for product cutouts and compositing

## Documentation
https://replicate.com/openai/gpt-image-1.5