# Kling O3 Image Image to Image - Core.Today AI API

> Kuaishou's Kling Omni 3 (O3) image-to-image via Fal.AI. Edits and transforms up to 10 reference images addressed in the prompt as @Image1 / @Element1, at 1K/2K/4K, as single images or a related series of 2-9, while preserving visual consistency.

- **Provider**: Kuaishou
- **Model ID**: fal-ai/kling-image/o3/image-to-image
- **Category**: Image Generation
- **Credits**: 66 per image (1K/2K 65, 4K 130 credits/image; series billed per image)
- **Speed**: Medium
- **Quality**: Ultra

## Features
- Up to 10 reference images addressed as @Image1, @Image2, ...
- Character elements (@Element1) for face/identity control
- 1K, 2K and 4K output
- Series mode for 2-9 related results

## Use Cases
- Turning sketches into photorealistic product renders
- Compositing products into new scenes
- Restyling photos while keeping the subject
- Consistent character variations from a reference

## 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 (max 2,500 characters). Reference the supplied images as @Image1, @Image2, ... (use @Image when there is only one) and character elements as @Element1, @Element2, ...
- **image_urls**: array - Reference images, 1-indexed in the prompt as @Image1, @Image2, ... Maximum 10.

### Optional
- **elements**: array - Characters or objects for face/identity control, referenced in the prompt as @Element1, @Element2, ... Each element takes a frontal_image_url and optional reference_image_urls.
- **resolution**: string (default: 1K) - Output resolution. 1K and 2K cost the same; 4K costs double per image.
  Options: 1K, 2K, 4K
- **result_type**: string (default: single) - 'single' generates num_images independent images; 'series' generates series_amount related images.
  Options: single, series
- **num_images**: integer (default: 1) - Number of images to generate when result_type is 'single' (1-9). Charged per image.
- **series_amount**: integer - Number of images in the series (2-9). Required when result_type is 'series'. Charged per image.
- **aspect_ratio**: string (default: auto) - Aspect ratio of the generated images. 'auto' picks one based on the input content.
  Options: auto, 16:9, 9:16, 1:1, 4:3, 3:4, 3:2, 2:3, 21:9
- **output_format**: string (default: png) - The format of the generated image.
  Options: jpeg, png, webp

## Examples

### Sketch to 3D render
Turn a product sketch into a photorealistic render at 1K.
```json
{
  "model": "fal-ai/kling-image/o3/image-to-image",
  "input": {
    "prompt": "Transform into photorealistic 3D render, add proper lighting and shadows, realistic materials and textures, maintain exact proportions and design from sketch, product visualization quality",
    "image_urls": [
      "https://v3b.fal.media/files/b/0a8d06d1/FTrwBsV9WI5twqMPkqvRh_19db9852c5ed4a109d81c52d27c9e433.png"
    ],
    "resolution": "1K",
    "aspect_ratio": "auto",
    "output_format": "png",
    "num_images": 1
  }
}
```

### 4K print render
The same transformation at 4K for print (double the per-image price).
```json
{
  "model": "fal-ai/kling-image/o3/image-to-image",
  "input": {
    "prompt": "Transform @Image into a studio product photo on a white seamless background with soft shadows",
    "image_urls": [
      "https://v3b.fal.media/files/b/0a8d06d1/FTrwBsV9WI5twqMPkqvRh_19db9852c5ed4a109d81c52d27c9e433.png"
    ],
    "resolution": "4K"
  }
}
```

## 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
- Refer to each input explicitly (@Image1, @Image2) so the model knows which image plays which role.
- 1K and 2K are the same price; 4K doubles it.
- In series mode you are charged for series_amount images (num_images is ignored).
- Keep 'auto' aspect ratio to follow the input image's framing.

## Documentation
https://fal.ai/models/fal-ai/kling-image/o3/image-to-image