# Clarity Upscaler - Core.Today AI API

> The famous creative upscaler (30M+ runs). Instead of just enlarging, it re-imagines detail while upscaling — with controllable creativity, resemblance, prompt guidance, and tiled diffusion for high scale factors.

- **Provider**: Clarity
- **Model ID**: philz1337x/clarity-upscaler
- **Category**: Image Generation
- **Credits**: 60 per generation
- **Speed**: Medium
- **Quality**: Ultra

## Features
- Creative detail re-imagination while upscaling
- Creativity and resemblance dials
- Prompt-guided enhancement
- Tiled diffusion for large outputs
- Optional hand fix and sharpening

## Use Cases
- Turning AI images into print-quality art
- Adding realistic texture to soft renders
- Skin and fabric detail enhancement
- Large-format poster preparation
- Upgrading old game assets

## 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
- **prompt**: string (default: masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>) - Guidance prompt (default 'masterpiece, best quality, highres')
- **negative_prompt**: string (default: (worst quality, low quality, normal quality:2) JuggernautNegative-neg) - Negative Prompt
- **scale_factor**: number (default: 2) - Upscale factor (higher factors take longer)
- **dynamic**: number (default: 6) - HDR intensity (1-50)
- **creativity**: number (default: 0.35) - How much new detail is invented (0-1). Higher deviates more from the original
- **resemblance**: number (default: 0.6) - How closely to follow the original (0-3)
- **tiling_width**: integer (default: 112) - Fractality, set lower tile width for a high Fractality
  Options: 16, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256
- **tiling_height**: integer (default: 144) - Fractality, set lower tile height for a high Fractality
  Options: 16, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256
- **sd_model**: string (default: juggernaut_reborn.safetensors [338b85bc4f]) - Stable Diffusion model checkpoint
  Options: epicrealism_naturalSinRC1VAE.safetensors [84d76a0328], juggernaut_reborn.safetensors [338b85bc4f], flat2DAnimerge_v45Sharp.safetensors
- **scheduler**: string (default: DPM++ 3M SDE Karras) - scheduler
  Options: DPM++ 2M Karras, DPM++ SDE Karras, DPM++ 2M SDE Exponential, DPM++ 2M SDE Karras, Euler a, Euler, LMS, Heun, DPM2, DPM2 a, DPM++ 2S a, DPM++ 2M, DPM++ SDE, DPM++ 2M SDE, DPM++ 2M SDE Heun, DPM++ 2M SDE Heun Karras, DPM++ 2M SDE Heun Exponential, DPM++ 3M SDE, DPM++ 3M SDE Karras, DPM++ 3M SDE Exponential, DPM fast, DPM adaptive, LMS Karras, DPM2 Karras, DPM2 a Karras, DPM++ 2S a Karras, Restart, DDIM, PLMS, UniPC
- **num_inference_steps**: integer (default: 18) - Number of denoising steps
- **seed**: integer (default: 1337) - Random seed for reproducible results
- **downscaling**: boolean (default: false) - Downscale the image before upscaling. Can improve quality and speed for images with high resolution but lower quality
- **downscaling_resolution**: integer (default: 768) - Downscaling resolution
- **lora_links**: string - Link to a lora file you want to use in your upscaling. Multiple links possible, seperated by comma
- **custom_sd_model**: string - custom_sd_model
- **sharpen**: number (default: 0) - Sharpening amount (0-10)
- **mask**: string - Mask image to mark areas that should be preserved during upscaling
- **handfix**: string (default: disabled) - Fix distorted hands ('disabled', 'hands_only', 'image_and_hands')
  Options: disabled, hands_only, image_and_hands
- **pattern**: boolean (default: false) - Upscale a pattern with seamless tiling
- **output_format**: string (default: png) - Format of the output images
  Options: webp, jpg, png

## Examples

### Print-Quality Art Upscale
2x upscale with moderate creative detail
```json
{
  "model": "philz1337x/clarity-upscaler",
  "input": {
    "image": "https://example.com/ai-art.png",
    "scale_factor": 2,
    "creativity": 0.35,
    "resemblance": 0.6
  }
}
```

## 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
- creativity 0.3-0.4 adds detail while staying faithful; 0.6+ visibly re-imagines the image
- Raise resemblance if faces or logos drift from the original
- Higher scale_factor increases runtime — 2x is the sweet spot for most work
- For pixel-faithful enlargement without re-imagination, use Real-ESRGAN instead
- Advanced controls (schedulers, LoRA links, tiling) are available via API

## Documentation
https://replicate.com/philz1337x/clarity-upscaler