Seedream 4.0
seedream-4.0
Upgraded image generation with improved lighting, visual detail, and complex prompt understanding.
Endpoints
Section titled “Endpoints”POST /v1/queue/seedream-4.0 # Create taskGET /v1/queue/seedream-4.0/requests/{request_id}/status # Check statusGET /v1/queue/seedream-4.0/requests/{request_id}/response # Get resultPUT /v1/queue/seedream-4.0/requests/{request_id}/cancel # Cancel taskRequest parameters
Section titled “Request parameters”curl --request 'POST' 'https://odirouter.ai/model/v1/queue/seedream-4.0' \ --header 'Authorization: Bearer YOUR_API_KEY' \ --header 'Content-Type: application/json' \ --location \ --data-raw '{ "prompt": "A clean studio shot of a red apple on a white table, soft natural light.", "image": "https://www.gstatic.com/webp/gallery/1.jpg", "sequential_image_generation": "disabled", "stream": false, "response_format": "url"}'Prompt for image generation, supports both Chinese and English
Recommended not to exceed 300 Chinese characters or 600 English words. Too many characters may cause information dispersion, and the model may ignore details, focusing only on key points, resulting in missing elements in the image
Input image information, supports URL or Base64 encoding, supports single or multiple image input
Image URL: Ensure the image URL is accessible
Base64 encoding: Follow this format data:image/<image_format>;base64,<Base64_encoding>. Note that <image_format> should be lowercase, e.g., data:image/png;base64,<base64_image>
Image requirements:
Supported formats: JPEG, PNG, WebP, BMP, TIFF, GIF
Aspect ratio (width/height) range: 1/16 to 16
Width and height (px) > 14
Size: not exceeding 10MB
Total pixels: not exceeding 6000x6000=36000000 px (limit on the pixel product of single image width and height, not on individual width or height values)
Maximum 14 reference images supported
Specify the size information of the generated image. Supports two methods, cannot be mixed
Method 1: Specify the resolution of the generated image, and describe the aspect ratio, shape, or usage in the prompt using natural language
Available values: 1K, 2K, 4K
Method 2: Specify the width and height pixel values
Total pixel range: 1280x720=921600 to 4096x4096=16777216
Aspect ratio range: 1/16 to 16
When using Method 2, both the total pixel range and aspect ratio range must be satisfied. Total pixels is a limit on the pixel product of single image width and height, not on individual width or height values
Valid example: 1600x600
Total pixel value 1600x600=960000, meets the range requirement [921600, 16777216]. Aspect ratio 1600/600=8/3, meets the range requirement [1/16, 16], so this example value is valid
Invalid example: 800x800
Total pixel value 800x800=640000, does not meet the minimum requirement 921600. Aspect ratio 800/800=1, although meets the range requirement [1/16, 16], but because it does not satisfy both constraints simultaneously, this example value is invalid
Recommended width and height pixel values:
1:1: 2048x2048
4:3: 2304x1728
3:4: 1728x2304
16:9: 2560x1440
9:16: 1440x2560
3:2: 2496x1664
2:3: 1664x2496
21:9: 3024x1296
Default: 2048x2048
Control whether to disable the image group feature
Image group: A set of content-related images generated based on your input
auto: Automatic mode, the model will determine whether to return image groups and the number of images based on the prompt
disabled: Disable image group feature, the model will only generate one image
Options: auto, disabled
Default: disabled
Configuration for image group feature. Only effective when sequential_image_generation is auto
Specify the maximum number of images that can be generated in this request
The actual number of images that can be generated is also affected by the number of input reference images. Number of input reference images + final generated images ≤ 15
Range: 1 to 15
Default: 15
Control whether to enable streaming output mode
false: Non-streaming output mode, returns all information at once after all images are generated
true: Streaming output mode, returns results for each image immediately. Streaming output mode is effective for both single image and image group scenarios
Options: true, false
Default: false
Specify the return format of the generated image
Generated images are in JPEG format, supports the following two return methods:
url: Returns image download link. Link is valid for 24 hours after image generation, please download images promptly
b64_json: Returns image data in Base64 encoded JSON format
Options: url, b64_json
Default: url
Whether to add watermark to the generated image
false: No watermark
true: Add "AI Generated" watermark in the bottom right corner
Options: true, false
Default: true
Configuration for prompt optimization feature
Set the mode for prompt optimization feature
standard: Standard mode, higher quality content generation, longer processing time
fast: Fast mode, shorter processing time, general quality
Options: standard, fast
Default: standard
Response parameters
Section titled “Response parameters”{ "request_id": "task_01hxyz...", "response_url": "https://odirouter.ai/model/v1/queue/seedream-4.0/requests/task_01hxyz.../response", "status_url": "https://odirouter.ai/model/v1/queue/seedream-4.0/requests/task_01hxyz.../status", "cancel_url": "https://odirouter.ai/model/v1/queue/seedream-4.0/requests/task_01hxyz.../cancel", "queue_position": 0}| Field | Type | Description |
|---|---|---|
request_id | string | Task ID used for all follow-up operations. |
response_url | string | URL to fetch the final result. |
status_url | string | URL to query task status. |
cancel_url | string | URL to cancel the task. |
queue_position | int | Current queue position. |