# Google Upscaler - Core.Today AI API

> Google's image upscaler (770K+ runs). Upscale images 2x or 4x using generative AI while preserving natural detail, with adjustable output compression — a simple, reliable enhancer at a flat price.

- **Provider**: Google
- **Model ID**: google/upscaler
- **Category**: Image Generation
- **Credits**: 46 per image (x2 and x4 same price)
- **Speed**: Fast
- **Quality**: High

## Features
- 2x or 4x upscaling with generative detail preservation
- Google-grade quality and consistency
- Adjustable output compression quality (1-100)
- Flat price regardless of upscale factor
- Simple 3-parameter API — easy to automate

## Use Cases
- Enhancing low-resolution product or listing photos
- Preparing AI-generated images for print
- Upscaling thumbnails into hero images
- Batch enhancement of archival photos
- Improving screenshots for documentation

## 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
- **image**: string - Input image to upscale

### Optional
- **upscale_factor**: string (default: x2) - Upscale factor: x2 or x4
  Options: x2, x4
- **compression_quality**: integer (default: 80) - Output compression quality (1-100)

## Examples

### 4x Upscale for Print
Maximum enlargement at the same flat price
```json
{
  "model": "google/upscaler",
  "input": {
    "image": "https://example.com/photo.jpg",
    "upscale_factor": "x4",
    "compression_quality": 95
  }
}
```

### Quick 2x Enhancement
Default settings for everyday enhancement
```json
{
  "model": "google/upscaler",
  "input": {
    "image": "https://example.com/thumbnail.jpg",
    "upscale_factor": "x2"
  }
}
```

## 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
- x2 and x4 cost the same 60 credits — pick x4 when you need the extra resolution
- Raise compression_quality to 90+ for print or archival output; the default 80 is fine for web
- For very large targets (up to 128MP) or factor-based control, compare with Pruna P-Image Upscale
- For face-specific restoration, GFPGAN or CodeFormer may fit better than a general upscaler

## Documentation
https://replicate.com/google/upscaler