# Seedream 5.0 Flash Edit - Core.Today AI API

> ByteDance Seedream 5.0 Flash image editing via Fal.AI. Fast, low-cost editing and multi-reference composition from up to 10 input images (referenced as Figure 1, Figure 2, ...), at 1K-2K output.

- **Provider**: ByteDance
- **Model ID**: bytedance/seedream/v5/flash/edit
- **Category**: Image Generation
- **Credits**: 63 per image (63 credits/image, same price at every size)
- **Speed**: Fast
- **Quality**: High

## Features
- Edit or combine up to 10 input images
- Refer to inputs as Figure 1, Figure 2, ... in the prompt
- Fast, budget-friendly — flat price per output image
- Automatic aspect ratio with auto_1K / auto_2K presets

## Use Cases
- Swapping products into existing marketing shots
- Applying logos and brand elements to mockups
- Background and style changes at scale
- Combining several references into one composition

## 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 - The text prompt used to edit the image. Refer to inputs as Figure 1, Figure 2, ...
- **image_urls**: array - Input reference images (up to 10; if more are sent, only the last 10 are used).

### Optional
- **image_size**: string (default: auto_2K) - Output size preset. auto_1K / auto_2K pick the aspect ratio automatically; total pixels are between 1024x1024 and 2048x2048. Same price for every size.
  Options: auto_2K, auto_1K, square_hd, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9
- **num_images**: integer (default: 1) - Number of separate generations to run, one image each (1-6). Charged per image.
- **output_format**: string (default: jpeg) - The file format of the generated image.
  Options: jpeg, png
- **enable_safety_checker**: boolean (default: true) - If set to true, the safety checker will be enabled.

## Examples

### Product swap with logo
Replace a product and render a logo onto it from three references.
```json
{
  "model": "bytedance/seedream/v5/flash/edit",
  "input": {
    "prompt": "Replace the product in Figure 1 with that in Figure 2. Seamlessly render the logo in Figure 3 into the product design, in a frosted glass texture.",
    "image_urls": [
      "https://storage.googleapis.com/falserverless/example_inputs/seedreamv45/seedream_v45_edit_input_1.png",
      "https://storage.googleapis.com/falserverless/example_inputs/seedreamv45/seedream_v45_edit_input_2.png",
      "https://storage.googleapis.com/falserverless/example_inputs/seedreamv45/seedream_v45_edit_input_3.png"
    ],
    "image_size": "auto_2K",
    "output_format": "jpeg"
  }
}
```

### Background change
Place the subject of a single photo on a new background, two variations.
```json
{
  "model": "bytedance/seedream/v5/flash/edit",
  "input": {
    "prompt": "Place the product from Figure 1 on a sunlit marble kitchen counter with soft morning light",
    "image_urls": [
      "https://storage.googleapis.com/falserverless/example_inputs/seedreamv45/seedream_v45_edit_input_2.png"
    ],
    "num_images": 2
  }
}
```

## 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
- Name each input by its position (Figure 1, Figure 2, ...) to say which image supplies what.
- Billing is per output image — input images add no charge.
- If more than 10 images are sent, only the last 10 are used.
- Keep auto_2K to follow the inputs' framing at the highest supported size.

## Documentation
https://fal.ai/models/bytedance/seedream/v5/flash/edit