> ## Documentation Index
> Fetch the complete documentation index at: https://docs.laozhang.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# SeeDream Image Generation/Editing

> SeeDream API: BytePlus Ark image model. Supports 4.0 (USD 0.035/image) and 4.5 (USD 0.045/image) versions, text-to-image and image-to-image, up to 10 reference images.

<Info>
  **API Endpoint:** `https://api2.laozhang.ai/v1/images/generations`

  **Available Models:**

  * `seedream-4-0-250828` - \$0.035/image
  * `seedream-4-5-251128` - \$0.045/image (Latest version)

  **Billing:** Pay-per-use
</Info>

<Warning>
  **Pricing Update Notice (January 2026)**

  Official subsidy has ended, SeeDream prices have been adjusted:

  * SeeDream 4.0: \$0.025/image → **\$0.035/image**
  * SeeDream 4.5 (New version): **\$0.045/image**
</Warning>

## Prerequisites

<Steps>
  <Step title="Get API Key">
    Log in to [laozhang.ai console](https://api2.laozhang.ai) to obtain your API key
  </Step>

  <Step title="Configure Billing Mode">
    Edit token settings and choose one of the following billing modes (same price for both):

    * **Volume Priority** (Recommended): Uses balance billing first, automatically switches when balance is insufficient. Suitable for most users
    * **Pay-per-call**: Direct deduction for each call. Suitable for strict budget control scenarios

    <Note>
      Both modes have **exactly the same price**, only the billing method differs. SeeDream 4.0 is \$0.035/image, 4.5 is \$0.045/image.
    </Note>

    <img src="https://mintcdn.com/laozhangai-edd05f2c/_loZ0Jy0ZI__xJ9z/images/sora2-token-setting.png?fit=max&auto=format&n=_loZ0Jy0ZI__xJ9z&q=85&s=d54128f51509467d6b73d207bbe5c86f" alt="Token Settings" width="1280" height="537" data-path="images/sora2-token-setting.png" />

    <Warning>
      If billing mode is not configured, API calls will fail. You must complete this configuration first!
    </Warning>
  </Step>
</Steps>

## Changelog

<Note>
  **September 11, 2025** - SeeDream 4.0 API launched officially, integrated by LaoZhang AI on the same day
</Note>

## Core Features

<CardGroup cols={2}>
  <Card title="Text-to-Image" icon="paintbrush">
    Generate high-quality images from text descriptions
  </Card>

  <Card title="Image-to-Image" icon="images">
    AI editing and redrawing based on source images
  </Card>

  <Card title="Multi-Image Input" icon="layers">
    Supports up to 10 reference images
  </Card>

  <Card title="Flexible Dimensions" icon="expand">
    Customizable image resolution and size
  </Card>

  <Card title="OpenAI Format" icon="code">
    Fully compatible with OpenAI Image API format
  </Card>

  <Card title="Flexible Versions" icon="tag">
    4.0 at \$0.035/image, 4.5 at \$0.045/image
  </Card>
</CardGroup>

## Key Information

### Source

<Tip>
  **Strategic Partnership**

  SeeDream 4.0 is powered by **BytePlus Ark** (international version). LaoZhang AI has established a strategic partnership to provide you with stable and reliable service.
</Tip>

### Technical Specifications

| Item                     | Description                                   |
| ------------------------ | --------------------------------------------- |
| **API Format**           | OpenAI Image API compatible                   |
| **Request Endpoint**     | `/v1/images/generations`                      |
| **Model Name**           | `seedream-4-0-250828` / `seedream-4-5-251128` |
| **Max Reference Images** | 10 images                                     |
| **Size Support**         | Customizable resolution                       |
| **Watermark**            | Optional (can be disabled)                    |

### Versions & Pricing

| Version          | Model ID              | Price         | Description                    |
| ---------------- | --------------------- | ------------- | ------------------------------ |
| **SeeDream 4.0** | `seedream-4-0-250828` | \$0.035/image | Stable version                 |
| **SeeDream 4.5** | `seedream-4-5-251128` | \$0.045/image | Latest version, higher quality |

<Note>
  💰 **Pricing Details**

  * **SeeDream 4.0**: \$0.035/image
  * **SeeDream 4.5**: \$0.045/image
  * **Billing note**: Pay-per-use pricing, with actual charges visible in call logs
</Note>

## Getting Started

<Steps>
  <Step title="Create Token">
    Login to [LaoZhang API Token Management](https://api2.laozhang.ai/token) and create a **pay-per-use** token

    <img src="https://mintcdn.com/laozhangai-edd05f2c/_loZ0Jy0ZI__xJ9z/images/token-create-per-request.png?fit=max&auto=format&n=_loZ0Jy0ZI__xJ9z&q=85&s=1fdceb804bb3ad50f83a21ec65b4d807" alt="Token Creation Interface" width="1364" height="1170" data-path="images/token-create-per-request.png" />
  </Step>

  <Step title="Select Billing Type">
    **Important**: Must select "Pay-per-use" type
  </Step>

  <Step title="Save Token">
    Copy the generated token (format: `sk-xxxxxx`) and replace `API_KEY` in the code
  </Step>
</Steps>

## Text-to-Image

### Python Version

```python theme={null}
#!/usr/bin/env python3
# -*- coding: utf-8 -*-

"""
Seedream API Image Generator - Pure Python Version
Just modify API_KEY to use
"""

# =================== Configuration Section ===================
API_KEY = "sk-"                                              # Replace with your API key
API_URL = "https://api2.laozhang.ai/v1/images/generations"    # API endpoint
PROMPT = "A beautiful sunset over mountains, realistic style"  # Image description
MODEL = "seedream-4-0-250828"                                # Model name
OUTPUT_DIR = "."                                             # Output directory
# ==============================================

import requests
import os
import datetime

def generate_image(prompt, output_dir="."):
    """Generate image and save to local"""
    print(f"🎨 Seedream Image Generator")
    print(f"📝 Prompt: {prompt}")
    print("=" * 50)
    
    # API request parameters
    headers = {
        "Content-Type": "application/json",
        "Authorization": f"Bearer {API_KEY}"
    }
    
    payload = {
        "model": MODEL,
        "prompt": prompt,
        "response_format": "url",
        "size": "2K",
        "watermark": False
    }
    
    try:
        print("⏳ Generating image...")
        
        # Call API
        response = requests.post(API_URL, headers=headers, json=payload, timeout=60)
        
        if response.status_code != 200:
            return False, f"API Error ({response.status_code}): {response.text}"
        
        # Parse response
        result = response.json()
        if "data" not in result or len(result["data"]) == 0:
            return False, "API returned no image data"
        
        # Get image URL
        image_url = result["data"][0]["url"]
        print(f"🌐 Got image URL successfully")
        
        # Download image
        print("⬇️ Downloading image...")
        img_response = requests.get(image_url, timeout=30)
        img_response.raise_for_status()
        
        # Generate filename with timestamp
        timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
        filename = os.path.join(output_dir, f"generated_{timestamp}.jpg")
        
        # Save image
        os.makedirs(output_dir, exist_ok=True)
        with open(filename, 'wb') as f:
            f.write(img_response.content)
        
        file_size = len(img_response.content)
        return True, f"✅ Generated successfully: {filename} ({file_size // 1024}KB)"
        
    except Exception as e:
        return False, f"Generation failed: {str(e)}"

def main():
    """Main function"""
    print("🚀 Seedream API Image Generator - Python Version")
    print("=" * 50)
    
    # Check API key
    if API_KEY == "sk-" or not API_KEY:
        print("⚠️  Please modify API_KEY at the top of the code")
        print("   Replace 'sk-' with your actual API key")
        print()
        print("📋 Instructions:")
        print("1. Modify API_KEY to your actual key")
        print("2. Optional: Modify PROMPT to generate different images")
        print("3. Run: python3 seedream-text-to-image.py")
        return
    
    # Execute image generation
    success, message = generate_image(PROMPT, OUTPUT_DIR)
    print(message)
    
    if success:
        print()
        print("🎉 Task completed!")
        print("💡 Tip: Modify PROMPT to generate different images")

if __name__ == "__main__":
    main()
```

### Curl Simple Version

```bash theme={null}
curl -X POST https://api2.laozhang.ai/v1/images/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-YOUR_API_KEY" \
  -d '{
    "model": "seedream-4-0-250828",
    "prompt": "A beautiful sunset over mountains, realistic style",
    "response_format": "url",
    "size": "2K",
    "watermark": false
}'
```

### Curl One-Line Version (with Auto Download)

<Tip>
  Requires python3 installation, automatically parses response and downloads image
</Tip>

```bash theme={null}
curl -X POST https://api2.laozhang.ai/v1/images/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-YOUR_API_KEY" \
  -d '{"model": "seedream-4-0-250828","prompt": "A beautiful sunset over mountains, realistic style","response_format": "url","size": "2K","watermark": false}' \
  | python3 -c "import json,sys,os; data=json.loads(sys.stdin.read()); os.system(f'curl -L -o generated_\$(date +%Y%m%d_%H%M%S).jpg \"{data[\"data\"][0][\"url\"]}\"') if 'data' in data else print('Error')"
```

## Image-to-Image

AI editing and redrawing based on source images, supports up to 10 reference images.

### Python Version

```python theme={null}
#!/usr/bin/env python3
# -*- coding: utf-8 -*-

"""
Seedream API Image Editor - Python Version
Generate new images based on source image, just modify API_KEY to use
"""

# =================== Configuration Section ===================
API_KEY = "sk-"                                              # Replace with your API key
API_URL = "https://api2.laozhang.ai/v1/images/generations"    # Image editing API endpoint
PROMPT = "Generate a close-up image of a dog lying on lush grass."  # Editing prompt
IMAGE_URL = "https://ark-doc.tos-ap-southeast-1.bytepluses.com/doc_image/seedream4_imageToimage.png"  # Source image URL
MODEL = "seedream-4-0-250828"                                # Model to use
OUTPUT_DIR = "."                                             # Output directory
# ==============================================

import requests
import os
import datetime

def edit_image(prompt, image_url, output_dir="."):
    """Generate edited image based on source image"""
    print(f"🎨 Seedream Image Editor")
    print(f"📝 Edit Prompt: {prompt}")
    print(f"🖼️  Source URL: {image_url[:50]}...")
    print("=" * 50)
    
    # API request parameters
    headers = {
        "Content-Type": "application/json",
        "Authorization": f"Bearer {API_KEY}"
    }
    
    payload = {
        "model": MODEL,
        "prompt": prompt,
        "image": image_url,
        "sequential_image_generation": "disabled",
        "response_format": "url",
        "size": "2K",
        "stream": False,
        "watermark": False
    }
    
    try:
        print("⏳ Editing image...")
        
        # Call API
        response = requests.post(API_URL, headers=headers, json=payload, timeout=60)
        
        if response.status_code != 200:
            return False, f"API Error ({response.status_code}): {response.text}"
        
        # Parse response
        result = response.json()
        if "data" not in result or len(result["data"]) == 0:
            return False, "API returned no image data"
        
        # Get image URL
        edited_image_url = result["data"][0]["url"]
        print(f"🌐 Got edited image URL successfully")
        
        # Download image
        print("⬇️ Downloading edited image...")
        img_response = requests.get(edited_image_url, timeout=30)
        img_response.raise_for_status()
        
        # Generate filename with timestamp
        timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
        filename = os.path.join(output_dir, f"edited_{timestamp}.jpg")
        
        # Save image
        os.makedirs(output_dir, exist_ok=True)
        with open(filename, 'wb') as f:
            f.write(img_response.content)
        
        file_size = len(img_response.content)
        return True, f"✅ Edit successful: {filename} ({file_size // 1024}KB)"
        
    except Exception as e:
        return False, f"Edit failed: {str(e)}"

def main():
    """Main function"""
    print("🚀 Seedream API Image Editor - Python Version")
    print("=" * 50)
    
    # Check API key
    if API_KEY == "sk-" or not API_KEY:
        print("⚠️  Please modify API_KEY at the top of the code")
        print("   Replace 'sk-' with your actual API key")
        print()
        print("📋 Instructions:")
        print("1. Modify API_KEY to your actual key")
        print("2. Modify IMAGE_URL to the image you want to edit")
        print("3. Modify PROMPT to describe the desired editing effect")
        print("4. Run: python3 seedream-image-edit.py")
        return
    
    # Execute image editing
    success, message = edit_image(PROMPT, IMAGE_URL, OUTPUT_DIR)
    print(message)
    
    if success:
        print()
        print("🎉 Edit completed!")
        print("💡 Tip: Modify PROMPT and IMAGE_URL to edit different images")

if __name__ == "__main__":
    main()
```

### Curl Version

```bash theme={null}
#!/bin/bash

# Seedream API Image Editor - Curl Version
# Generate new images based on source image, modify API_KEY to use

# =============== Configuration Section ===============
API_KEY="sk-YOUR_API_KEY"                                    # Replace with your API key
API_URL="https://api2.laozhang.ai/v1/images/generations"      # Image editing API endpoint
PROMPT="Generate a close-up image of a dog lying on lush grass."  # Editing prompt
IMAGE_URL="https://ark-doc.tos-ap-southeast-1.bytepluses.com/doc_image/seedream4_imageToimage.png"  # Source image URL
MODEL="seedream-4-0-250828"                                  # Model name
# ====================================

# Check API key
if [ "$API_KEY" = "sk-YOUR_API_KEY" ]; then
    echo "⚠️  Please modify API_KEY in the script"
    echo "   Replace 'sk-YOUR_API_KEY' with your actual API key"
    echo ""
    echo "📋 Instructions:"
    echo "1. Modify API_KEY to your actual key"
    echo "2. Modify IMAGE_URL to the image you want to edit"
    echo "3. Modify PROMPT to describe the desired editing effect"
    echo "4. Run: ./seedream-image-edit.sh"
    exit 1
fi

# Generate output filename
TIMESTAMP=$(date +"%Y%m%d_%H%M%S")
OUTPUT_FILE="edited_${TIMESTAMP}.jpg"

echo "🎨 Seedream API Image Editor - Curl Version"
echo "========================================"
echo "📝 Edit Prompt: $PROMPT"
echo "🖼️  Source URL: ${IMAGE_URL:0:50}..."
echo "📁 Output File: $OUTPUT_FILE"
echo "========================================"

# Step 1: Call API to get JSON response
echo "⏳ Editing image..."

# Build JSON request body
JSON_DATA=$(cat <<EOF
{
  "model": "$MODEL",
  "prompt": "$PROMPT",
  "image": "$IMAGE_URL",
  "sequential_image_generation": "disabled",
  "response_format": "url",
  "size": "2K",
  "stream": false,
  "watermark": false
}
EOF
)

RESPONSE=$(curl -s -X POST "$API_URL" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $API_KEY" \
  -d "$JSON_DATA")

# Check API response
if [ -z "$RESPONSE" ]; then
    echo "❌ No API response, please check network connection"
    exit 1
fi

echo "✅ API response successful"

# Step 2: Parse JSON with Python and extract image URL
echo "🔍 Parsing edited image URL..."
EDITED_IMAGE_URL=$(echo "$RESPONSE" | python3 -c "
import json
import sys

try:
    data = json.loads(sys.stdin.read())
    
    if 'data' in data and len(data['data']) > 0:
        url = data['data'][0]['url']
        print(url)
    else:
        print('ERROR: Image URL not found')
        sys.exit(1)
        
except Exception as e:
    print(f'ERROR: {e}')
    sys.exit(1)
")

# Check URL parsing result
if [[ "$EDITED_IMAGE_URL" == ERROR* ]]; then
    echo "❌ URL parsing failed: $EDITED_IMAGE_URL"
    echo "📄 Original API response:"
    echo "$RESPONSE"
    exit 1
fi

echo "🌐 Edited image URL retrieved successfully"

# Step 3: Download edited image
echo "⬇️ Downloading edited image..."
curl -s -L -o "$OUTPUT_FILE" "$EDITED_IMAGE_URL"

# Check download result
if [ -f "$OUTPUT_FILE" ] && [ -s "$OUTPUT_FILE" ]; then
    FILE_SIZE=$(du -h "$OUTPUT_FILE" | cut -f1)
    echo "✅ Image edited successfully!"
    echo "📁 File: $OUTPUT_FILE"
    echo "📊 Size: $FILE_SIZE"
    echo ""
    echo "🎉 Complete! Check the edited image file"
    echo "💡 Tip: Modify PROMPT and IMAGE_URL to edit different images"
else
    echo "❌ Image download failed"
    echo "🔗 You can manually access: $EDITED_IMAGE_URL"
fi
```

## Resources

### Official Documentation

<CardGroup cols={2}>
  <Card title="API Documentation" icon="book" href="https://docs.byteplus.com/en/docs/ModelArk/1541523">
    BytePlus ModelArk official technical documentation
  </Card>

  <Card title="User Manual" icon="file-text" href="https://docs.byteplus.com/en/docs/ModelArk/seedream-4-0-user-manual">
    Seedream 4.0 User Manual
  </Card>

  <Card title="User Guide" icon="compass" href="https://docs.byteplus.com/en/docs/ModelArk/seedream-4-0-user-guide">
    Seedream 4.0 User Guide
  </Card>

  <Card title="Prompting Tips" icon="lightbulb" href="https://docs.byteplus.com/en/docs/ModelArk/seedream-4-0-user-manual#prompting-guide">
    Prompting and style keyword guide
  </Card>
</CardGroup>

### Common Questions

<AccordionGroup>
  <Accordion title="Dimensions and Resolution" icon="ruler">
    Supports custom image dimensions via `size` parameter:

    * `2K`: 2048 pixels (recommended)
    * Also supports custom resolutions

    See official documentation for parameter details.
  </Accordion>

  <Accordion title="Reference Image Count" icon="images">
    * Text-to-Image: No reference images needed
    * Image-to-Image: Supports 1-10 reference images
    * Pass via `image` parameter as URL or Base64
  </Accordion>

  <Accordion title="Watermark Settings" icon="droplet">
    Control via `watermark` parameter:

    * `false`: No watermark (recommended)
    * `true`: With watermark
  </Accordion>

  <Accordion title="Response Format" icon="file-code">
    Supports two response formats (`response_format`):

    * `url`: Returns image link (recommended, convenient for download)
    * `b64_json`: Returns Base64 encoding
  </Accordion>

  <Accordion title="Generation Mode" icon="sliders-horizontal">
    Control via `sequential_image_generation` parameter:

    * `disabled`: Standard generation mode (recommended)
    * `enabled`: Sequential generation mode
  </Accordion>
</AccordionGroup>

## API Parameter Reference

### Request Parameters

| Parameter                     | Type    | Required | Description                                         |
| ----------------------------- | ------- | -------- | --------------------------------------------------- |
| `model`                       | string  | ✓        | Model name: `seedream-4-0-250828`                   |
| `prompt`                      | string  | ✓        | Image description or editing prompt                 |
| `image`                       | string  | ✗        | Reference image URL (for image-to-image)            |
| `size`                        | string  | ✗        | Image size, default `2K`                            |
| `response_format`             | string  | ✗        | Response format: `url` or `b64_json`, default `url` |
| `watermark`                   | boolean | ✗        | Whether to add watermark, default `false`           |
| `sequential_image_generation` | string  | ✗        | Generation mode, default `disabled`                 |
| `stream`                      | boolean | ✗        | Whether to stream response, default `false`         |

### Response Format

```json theme={null}
{
  "created": 1726051200,
  "data": [
    {
      "url": "https://example.com/generated_image.jpg",
      "revised_prompt": "..."
    }
  ]
}
```

## Next Steps

<CardGroup cols={2}>
  <Card title="Nano Banana Image Generation" icon="sparkles" href="/en/api-capabilities/nano-banana-image">
    \$0.025/image Gemini image generation
  </Card>

  <Card title="FLUX Image Generation" icon="bolt" href="/en/api-capabilities/flux-image-generation">
    Check out FLUX image generation API
  </Card>

  <Card title="API Console" icon="key" href="https://api2.laozhang.ai/token">
    Create API token
  </Card>

  <Card title="Billing Console" icon="wallet" href="https://api2.laozhang.ai">
    Review balance, account credit updates, and call logs
  </Card>
</CardGroup>