Generations（Seedream 4.0）

POST

/doubao/images/generations

The latest Seedream model supports both text-to-image and image-to-image generation.

Model Name:
doubao-seedream-4-0-250828

Features:

Generate image sets (image set: a group of content-related images generated based on your input; requires sequential_image_generation to be set to auto)

Multi-image to image set: Generate a group of content-related images based on your input of multiple reference images (2-10) + text prompts (the number of reference images input + the number of final generated images ≤ 15).

Single-image to image set: Generate a group of content-related images based on your input of a single reference image + text prompts (up to 14 images can be generated).

Text to image set: Generate a group of content-related images based on your input of text prompts (up to 15 images can be generated).

Generate single image (set sequential_image_generation to disabled)

Multi-image to image: Generate a single image based on your input of multiple reference images (2-10) + text prompts.

Single-image to image: Generate a single image based on your input of a single reference image + text prompts.

Text to image: Generate a single image based on your input of text prompts.

Price: 0.03 PTC/image

Resolution Level	Aspect Ratio	Width (px)	Height (px)
1K	1:1	1024	1024
	4:3	1152	864
	3:4	864	1152
	16:9	1280	720
	9:16	720	1280
	3:2	1248	832
	2:3	832	1248
	21:9	1512	648
2K	1:1	2048	2048
	4:3	2304	1728
	3:4	1728	2304
	16:9	2560	1440
	9:16	1440	2560
	3:2	2496	1664
	2:3	1664	2496
	21:9	3024	1296
4K	1:1	4096	4096
	4:3	4736	3552
	3:4	3552	4736
	16:9	5472	2072
	9:16	2072	5472
	3:2	5024	3360
	2:3	3360	5024
	21:9	6272	2688

Request

Header Params

Body Params application/json

Examples

图生图文生图

{
    "model": "doubao-seedream-4-0-250828",
    "prompt": "生成3张女孩和奶牛玩偶在游乐园开心地坐过山车的图片，涵盖早晨、中午、晚上",
    "image": [
        "https://ark-project.tos-cn-beijing.volces.com/doc_image/seedream4_imagesToimages_1.png",
        "https://ark-project.tos-cn-beijing.volces.com/doc_image/seedream4_imagesToimages_2.png"
    ],
    "sequential_image_generation": "auto",
    "sequential_image_generation_options": {
        "max_images": 3
    },
    "response_format": "url",
    "size": "2K",
    "stream": false,
    "watermark": false
}

Request Code Samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

cURL

图生图文生图

curl --location --request POST 'https://api.302.ai/doubao/images/generations' \
--header 'Content-Type: application/json' \
--data-raw '{
    "model": "doubao-seedream-4-0-250828",
    "prompt": "生成3张女孩和奶牛玩偶在游乐园开心地坐过山车的图片，涵盖早晨、中午、晚上",
    "image": [
        "https://ark-project.tos-cn-beijing.volces.com/doc_image/seedream4_imagesToimages_1.png",
        "https://ark-project.tos-cn-beijing.volces.com/doc_image/seedream4_imagesToimages_2.png"
    ],
    "sequential_image_generation": "auto",
    "sequential_image_generation_options": {
        "max_images": 3
    },
    "response_format": "url",
    "size": "2K",
    "stream": false,
    "watermark": false
}'

Responses

🟢200成功

application/json

Response Parameter Description

Field List

Field Name	Type	Description
model	string	The model ID used for this request (model name-version).
created	integer	The Unix timestamp (seconds) when this request was created.
data	list	Information about the output images, including the image download URL or Base64. - When the return format is specified as url, the field is `url` - When the return format is specified as b64_json, the field is `b64_json` Note: To ensure information security, the url link will expire within 24 hours after generation. Please make sure to save the image in time.
usage	object	Information related to image generation. - `usage.generated_images` (integer): Number of images generated by the model.
error	object	Error information (optional). - `error.code` (string): Error code - `error.message` (string): Error message

data Field Description

When response_format is url, each element in the data array contains the field:

url: Image download address

When response_format is b64_json, each element in the data array contains the field:

b64_json: Base64 encoded string of the image

usage Field Description

usage.generated_images: Number of images generated by the model (integer)

error Field Description (Optional)

error.code: Error code (string)

error.message: Error message (string)

Body

Example

{
    "model": "doubao-seedream-4-0-250828",
    "created": 1757321139,
    "data": [
        {
            "url": "https://...",
            "size": "3104x1312"
        }
    ],
    "usage": {
        "generated_images": 1,
        "output_tokens": xxx,
        "total_tokens": xxx
    }
}

Modified at 2025-09-09 10:01:30

Luma-Photon-Flash（Fast image generation）

Generations (Jimeng Image Generation)