# Z-Image Turbo - Core.Today AI API

> Pruna's ultra-fast Z-Image Turbo (48M+ runs). Megapixel-priced image generation up to 2048x2048 — pay exactly for the resolution you generate, with excellent price/performance for high-volume use.

- **Provider**: Pruna AI
- **Model ID**: prunaai/z-image-turbo
- **Category**: Image Generation
- **Credits**: 40 per megapixel (1024x1024 = 40, 2048x2048 = 160, 512x512 = 10)
- **Speed**: Fast
- **Quality**: High

## Features
- Ultra-fast turbo generation (8 steps)
- Megapixel-based pricing — pay for what you generate
- Up to 2048x2048 output
- go_fast mode for extra speed
- 48M+ runs and climbing fast

## Use Cases
- High-volume image generation
- Small thumbnails at fractional cost
- Rapid prompt iteration
- Cost-optimized production pipelines
- 2K wallpapers and banners

## 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 - Text prompt for image generation

### Optional
- **height**: integer (default: 1024) - Output height (64-2048). Credits scale with width x height
- **width**: integer (default: 1024) - Output width (64-2048). Credits scale with width x height
- **num_inference_steps**: integer (default: 8) - Inference steps (1-50). 8 is the turbo sweet spot
- **guidance_scale**: number (default: 0) - Guidance scale. Should be 0 for turbo models
- **seed**: integer - Random seed for reproducible generation
- **go_fast**: boolean (default: false) - Apply additional optimizations for faster generation
- **output_format**: string (default: jpg) - Output image format
  Options: png, jpg, webp
- **output_quality**: integer (default: 80) - Quality when saving the output images, from 0 to 100. 100 is best quality, 0 is lowest quality. Not relevant for .png outputs

## Examples

### Fast 1MP Generation
Standard 1024x1024 image for 30 credits
```json
{
  "model": "prunaai/z-image-turbo",
  "input": {
    "prompt": "A cozy coffee shop interior, warm lighting, photorealistic",
    "width": 1024,
    "height": 1024
  }
}
```

### Cheap Thumbnail Batch
512x512 thumbnails at 7.5 credits each
```json
{
  "model": "prunaai/z-image-turbo",
  "input": {
    "prompt": "Minimalist app icon, gradient background, flat design",
    "width": 512,
    "height": 512
  }
}
```

## 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
- Credits scale with resolution: 512px = 7.5, 1024px = 30, 2048px = 120 — size images to what you need
- Keep guidance_scale 0 and steps 8; that's what turbo is tuned for
- Enable go_fast for maximum throughput in bulk pipelines
- Generate small for drafts, then regenerate the winner at 2048px with the same seed

## Documentation
https://replicate.com/prunaai/z-image-turbo