seedream series
Seedream Series Unified Endpoint Guide
- This page explains how
POST /v1/images/generationsbehaves for the seedream series. - Focus is on unified-endpoint parameter behavior, not only surface field names.
Basic Information
| Item | Description |
|---|---|
| Unified endpoint | POST /v1/images/generations |
| Auth | Authorization: Bearer YOUR_API_KEY |
| Models | doubao-seedream-4.0, doubao-seedream-4.5, bytedance-seedream-5.0 |
Prepare these variables:
BASE_URL="https://api-platform.ope.ai"
API_KEY="YOUR_API_KEY"
For pricing, see Model Pricing.
1. Unified Field Behavior
| Unified field | Platform handling | Notes |
|---|---|---|
model | Normalization | Seedream aliases are normalized (see table below) |
prompt | Direct handling | Used as the generation prompt |
size | Direct handling | Default is applied when omitted |
response_format | Validation | Only b64_json and url are accepted |
image | Direct handling | Supports image-to-image / multi-image blend |
stream | Default | Default false |
watermark | Default | Default false |
sequential_image_generation | Conditional | Seedream-only, allowed: disabled, auto |
n | Conditional rewrite | When seedream + n > 1 + sequential_image_generation omitted, auto rewritten to auto + max_images=n |
Fields currently ignored on this unified path: quality, style, user, output_format, output_compression, background, moderation.
2. Model ID Normalization (Platform Internal)
Request model | Normalized model |
|---|---|
doubao-seedream-4.5 / doubao-seedream-4-5-251128 | seedream-4-5-251128 |
doubao-seedream-4.0 / doubao-seedream-4-0-250828 | seedream-4-0-250828 |
bytedance-seedream-5.0 | bytedance-seedream-5.0 |
size defaults:
doubao-seedream-3-0-t2i-250415/seedream-3-0-t2i-250415:1024x1024- Other seedream models:
2K
3. Supported size Values (Seedream Native Specs)
sizeis passed through. You can send2K/3K/4K, or send explicit resolutions (recommended).
Reference: Volcengine official docs Image Generation
| Tier | Ratio | Resolution (size) |
|---|---|---|
| 2K | 1:1 | 2048x2048 |
| 2K | 4:3 | 2304x1728 |
| 2K | 3:4 | 1728x2304 |
| 2K | 16:9 | 2848x1600 |
| 2K | 9:16 | 1600x2848 |
| 2K | 3:2 | 2496x1664 |
| 2K | 2:3 | 1664x2496 |
| 2K | 21:9 | 3136x1344 |
| 3K | 1:1 | 3072x3072 |
| 3K | 4:3 | 3456x2592 |
| 3K | 3:4 | 2592x3456 |
| 3K | 16:9 | 4096x2304 |
| 3K | 9:16 | 2304x4096 |
| 3K | 2:3 | 2496x3744 |
| 3K | 3:2 | 3744x2496 |
| 3K | 21:9 | 4704x2016 |
| 4K | 1:1 | 4096x4096 |
| 4K | 3:4 | 3520x4704 |
| 4K | 4:3 | 4704x3520 |
| 4K | 16:9 | 5504x3040 |
| 4K | 9:16 | 3040x5504 |
| 4K | 2:3 | 3328x4992 |
| 4K | 3:2 | 4992x3328 |
| 4K | 21:9 | 6240x2656 |
4. n Semantics (Important)
ndefaults to1.- For seedream,
nis not a strict "generate exactly n images". - In sequential mode,
nis only used asmax_images; actual generated count is decided by the model based on the prompt. - Recommended official limit is
[1, 15]; keep requests within this range.
Gateway auto-rewrite example (when n > 1 and sequential_image_generation is omitted):
{
"sequential_image_generation": "auto",
"sequential_image_generation_options": {
"max_images": 8
}
}
5. Create Request Example
curl --request POST "$BASE_URL/v1/images/generations" \
--header "Authorization: Bearer $API_KEY" \
--header "Content-Type: application/json" \
--data '{
"model": "bytedance-seedream-5.0",
"prompt": "Generate a cohesive set of cinematic posters of the same character from different camera angles.",
"size": "4096x2304",
"n": 8,
"response_format": "url"
}'