Seedream 4.0 API

Seedream 4.0 is ByteDance Seed's image model that unifies generation and editing, prioritizes reference consistency, and supports up to 4K output.

Next generation available: Seedream 5.0 with Deep Thinking, Web Search, and 14 reference images. Upgrade to Seedream 5.0 →

Prompt

A beautiful landscape with mountains and a lake, golden hour lighting, photorealistic style

Parameters

size

image_urls

Estimated Cost (per image)

1.8 Credits

Total = Base Price × Count (n)

Sample Result

Prompt

91 (suggested: 2,000)

Input Images

Upload reference images

Click to upload or drag and drop

Supported formats: JPG, JPEG, PNG, WEBP
Maximum file size: 10MB; Maximum files: 10

Size

Quality

Count

Specifies the maximum number of images to generate (1–15).

History

Max 20 items

0 running · 0 completed

Your generation history will appear here

Seedream 4.0 API for high‑consistency image generation and editing

Seedream 4.0 is a image creation model from ByteDance’s Seed team that unifies image generation and editing in a single architecture. It emphasizes reference consistency and can produce high‑definition images up to 4K.

Pricing

Model	Mode	Price
Seedream 4	Image Generation	Popular $0.0222/ image(1.6 Credits)

Popular

Seedream 4

Image Generation

Price:

$0.0222/ image

(1.6 Credits)

If it's down, we automatically use the next cheapest available—ensuring 99.9% uptime at the best possible price.

What can you build with Seedream 4.0 API?

Batch image generation for campaigns

Generate multiple variants in one request using the `n` parameter (1–15). This is ideal for ad sets, hero images, and product variants where you want controlled diversity without rebuilding prompts for each version.

Start using Seedream 4.0 API

Multi-reference image editing

Upload multiple reference images via `image_urls` to guide image-to-image edits and preserve key identity elements such as logos, materials, or subjects. This is useful for product catalogs, brand assets, and batch edits.

Try image-to-image edits

Resolution control for production output

Seedream 4.0 API supports simplified sizing (1K/2K/4K) and explicit pixel sizes within the documented range. Use this to match ad specs, landing pages, and marketplace requirements without manual resizing.

Generate production-ready sizes

Why developers choose Seedream 4.0 API

A unified, async image API that balances batch productivity, reference-guided consistency, and predictable sizing.

Unified generation and editing

Run text-to-image, image-to-image, and image editing through one endpoint and the same authentication flow.

Reference‑guided consistency

Use multi-reference inputs to keep identity, branding, or composition stable across outputs and variations.

Async workflows built for scale

Tasks return a task ID and can trigger a callback URL when complete, enabling reliable pipelines and queue-based processing.

How to integrate Seedream 4.0 API

A simple, async workflow designed for production automation.

Step 1 — Create an API key

Generate your evolink.ai API key in the dashboard and use Bearer token authentication for all requests.

Step 2 — Call /v1/images/generations

Send `model: "doubao-seedream-4.0"`, your prompt, and optional parameters such as `n`, `size`, `image_urls`, `prompt_priority`, and `callback_url`.

Step 3 — Retrieve results

Poll the task status endpoint or receive a webhook callback. Save returned image URLs promptly (links are valid for 24 hours).

Open Seedream 4.0 API docs

Seedream 4.0 API capabilities

Documented behavior and limits that matter for production teams.

Modes

Text-to-image + image-to-image

One endpoint supports generation from text and edits based on reference images.

Batching

Batch outputs up to 15

Use `n` to request multiple candidates per call, with total reference + output images capped at 15.

Consistency

Multi-reference inputs

Provide up to 14 reference images via `image_urls` to guide edits and preserve identity.

Resolution

Flexible sizing

Pick simplified sizes (1K/2K/4K) or explicit pixel dimensions within the supported range.

Automation

Async task workflow

Each request creates a task ID, enabling queue-based processing and optional callbacks.

Compatibility

Broad input formats

Reference images can use common formats like JPG, PNG, WebP, BMP, TIFF, and GIF.

Seedream 4.0 FAQs

Everything you need to know about the product and billing.

Seedream 4.0 is ByteDance Seed’s next‑generation image creation model. It unifies image generation and image editing in a single architecture, emphasizes reference consistency, and supports output up to 4K.

Seedream 4.0 accepts text, images, or any combination as input, supporting text‑to‑image generation, image‑to‑image translation, single‑image editing, multi‑image editing, and image composition.

It supports multi‑image input and subject‑consistent fusion, helping preserve key identity details across multiple outputs.

It can produce high‑definition images up to 4K, which is useful for production‑ready visuals and large‑format use cases.

Yes. ByteDance notes it can switch between distinct artistic styles (for example, Baroque to Cyberpunk) and even blend styles to create new visual looks.

ByteDance reports that Seedream 4.0 delivers much faster inference than its predecessor, helping teams iterate more quickly.

Yes. Reference consistency is a core capability, helping the model maintain stable subjects and details across related images.

Seedream 4.0 is officially available per ByteDance Seed’s release, and BytePlus provides the official ModelArk documentation.

Explore more image APIs on EvoLink

Seedream 4.5|Seedream 5.0 Lite

API Reference

Select endpoint

Authentication

All APIs require Bearer Token authentication.

Header

Authorization: 
Bearer YOUR_API_KEY

Get API Key

POST

/v1/images/generations

Generate Image

Create an image generation task using text prompts or reference images. Supports text-to-image, image-to-image, and image editing modes.

Asynchronous processing mode, use the returned task ID to .

Generated image links are valid for 24 hours, please save them promptly.

Request Parameters

modelstringRequiredDefault: doubao-seedream-4.0

Image generation model name.

Exampledoubao-seedream-4.0

promptstringRequired

Prompt describing the image to be generated or how to edit the input image.

Notes

Limited to 2000 tokens

ExampleA cat playing in the grass

sizestringOptionalDefault: auto

Aspect ratio of generated images.

Value	Description
auto	Auto - model determines optimal ratio
1:1	Square
2:3 / 3:2	Portrait / Landscape
3:4 / 4:3	Portrait / Landscape
4:5 / 5:4	Portrait / Landscape
9:16 / 16:9	Vertical / Horizontal
21:9	Ultra-wide

Notes

Backend auto-converts ratio + quality to exact pixels

Example16:9

qualitystringOptionalDefault: 2K

Resolution tier for generated images. Backend converts ratio + quality to exact pixel dimensions.

Value	Description
1K	Standard quality
2K	High quality (default)
4K	Ultra quality (higher cost)

Example2K

nintegerOptionalDefault: 1

Number of images to generate.

Value	Description
1-15	Batch generation supported

Example1

image_urlsarrayOptional

Reference image URL list for image-to-image and image editing functions.

Notes

Max 10 images per request
Max size: 10MB per image
Formats: .jpeg, .jpg, .png
Aspect ratio range: [1/3, 3]
Total pixels: ≤6000×6000
URLs must be directly accessible by the server

Examplehttps://example.com/image1.png

prompt_priorityenum<string>OptionalDefault: standard

Prompt optimization strategy for setting the prompt optimization mode.

Value	Description
standard	Standard mode, higher quality, slower
fast	Fast mode, faster speed, normal quality

Examplestandard

callback_urlstringOptional

HTTPS callback address after task completion.

Notes

Triggered on completion, failure, or cancellation
Sent after billing confirmation
HTTPS only, no internal IPs
Max length: 2048 chars
Timeout: 10s, Max 3 retries

Examplehttps://your-domain.com/webhooks/image-task-completed

Request Example

{
  "model": "doubao-seedream-4.0",
  "prompt": "A cat playing on the grass",
  "size": "16:9",
  "quality": "2K"
}

Response Example

{
  "created": 1757165031,
  "id": "task-unified-1757165031-uyujaw3d",
  "model": "doubao-seedream-4.0",
  "object": "image.generation.task",
  "progress": 0,
  "status": "pending",
  "task_info": {
    "can_cancel": true,
    "estimated_time": 45
  },
  "type": "image",
  "usage": {
    "billing_rule": "per_call",
    "credits_reserved": 1.8,
    "user_group": "default"
  }
}