Seedream 5.5 API
ByteDance Seedream 5.5 — photographic realism and strong visual aesthetics, consistent multi-round editing that keeps unedited areas steady, and accurate small and multilingual text. Native transparent PNG output, a 3K resolution tier, and a layer decomposition endpoint that splits one image into a base image plus transparent layers.
Three endpoints on one family: text-to-image, Edit with 1 to 10 reference images, and Layer Decomposition, which turns one image into a base image plus transparent PNG layers. Text-to-image and Edit take 1k / 1.5k / 2k / 3k output, fifteen aspect ratios from 1:3 to 3:1 (including 9:21 and 21:9), png or jpeg output, and a background control (auto / transparent / opaque) that returns a PNG with an alpha channel straight from the model.
/filters:quality(82)/examples/f5f78e1cf4e0a817381598e24db55ca1/f1aef82f-920b-4df5-b513-9878f69c7d52-92bbc258f913.webp)
Overview
About the Seedream 5.5 API
What Seedream 5.5 does, how it fits in the ByteDance model lineup, and why teams reach for it.
Seedream 5.5 is a image generation and editing model from ByteDance, available through the WaveSpeedAI REST API. ByteDance Seedream 5.5 — photographic realism and strong visual aesthetics, consistent multi-round editing that keeps unedited areas steady, and accurate small and multilingual text. Native transparent PNG output, a 3K resolution tier, and a layer decomposition endpoint that splits one image into a base image plus transparent layers.
Three endpoints on one family: text-to-image, Edit with 1 to 10 reference images, and Layer Decomposition, which turns one image into a base image plus transparent PNG layers. Text-to-image and Edit take 1k / 1.5k / 2k / 3k output, fifteen aspect ratios from 1:3 to 3:1 (including 9:21 and 21:9), png or jpeg output, and a background control (auto / transparent / opaque) that returns a PNG with an alpha channel straight from the model.
The Seedream 5.5 family on WaveSpeedAI ships 3 REST endpoints covering Image-Tools, Text-To-Image, Image-To-Image workflows. Each variant carries its own pricing, parameter knobs, and example outputs — pick the one that matches your input modality and production constraints, or call several from the same API key to compose multi-step pipelines.
Run Seedream 5.5 through the same API key, billing account, and rate-limit envelope you use for the other 1,000+ AI models on WaveSpeedAI. No separate vendor setup, no per-provider SDKs, no per-vendor rate-limit envelopes — one integration covers everything from text-to-image and text-to-video through audio synthesis, 3D generation, upscaling, and editing.
Specs
Seedream 5.5 API capabilities and release status
The model-specific details developers search for before choosing an API: availability, expected output length, reference support, and the current live fallback.
Endpoints
3 variants
Text-to-image, Edit (1 to 10 reference images), and Layer Decomposition.
Resolution
1k / 1.5k / 2k / 3k
3k is about 8 megapixels (3840x2160 at 16:9). Layer Decomposition takes 1k, 1.5k, and 2k.
Transparent background
Native PNG alpha
background = auto, transparent, or opaque on text-to-image and Edit; transparent needs png output.
Layers
Up to 99
max_images from 2 to 100 total outputs (30 by default): a base image plus transparent PNG layers.
Aspect ratios
15 options
1:1, 1:2, 2:1, 1:3, 3:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 9:21, and 21:9.
Endpoints
All Seedream 5.5 API endpoints
3 Seedream 5.5 endpoints available now on WaveSpeedAI — pick the variant that matches your workflow.
/filters:quality(82)/media/images/8c8fcb40-841c-4a9c-8760-9e682fb7c494.webp)
Seedream v5.5 Layer Decomposition
Seedream 5.5 Layer Decomposition separates a single image into a base image and transparent layers, enabling flexible compositing, image editing, asset extraction, and layered design workflows. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
/filters:quality(82)/media/images/9fe24f82-29bb-48e1-b20b-4fd74e0642e6.webp)
Seedream v5.5
Seedream 5.5 Text to Image by ByteDance generates images from text prompts with improved photographic realism and aesthetics, native transparent-background output, aspect ratio selection, and 1K, 1.5K, 2K and 3K output tiers. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
/filters:quality(82)/media/images/c48b23e9-ca36-4f8c-b525-5f3ba60ce4f0.webp)
Seedream v5.5 Edit
Seedream 5.5 Edit by ByteDance edits and generates images from single-image or multi-reference inputs, supporting up to 10 reference images, consistent multi-turn edits, native transparent-background output, and 1K, 1.5K, 2K and 3K output tiers. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Examples
See Seedream 5.5 in action
Real outputs generated by the Seedream 5.5 API. Hover any video to preview, click to open the full-size viewer.
How to
How to use the Seedream 5.5 API
Four steps from signup to a finished generation. Full Python, Node.js, and cURL examples are in the API section below.
- 01
Get an API key
Sign up for a WaveSpeedAI account and copy your API key from the dashboard. New accounts come with free starter credits — enough to run the playground a few dozen times before billing kicks in.
- 02
Submit a prediction
POST your input as JSON to https://api.wavespeed.ai/api/v3/bytedance/seedream-v5.5. The endpoint returns a prediction id immediately — generations are async so you don't hold an open connection during inference.
- 03
Poll for completion
GET https://api.wavespeed.ai/api/v3/predictions/{request_id}/result. Return outputs on completed; stop with an error on failed, cancelled, timeout, or deleted; continue polling for every other status.
- 04
Read the output URL
Once status is "completed", read the URL from data.outputs[0]. The URL points to your generated media on the WaveSpeedAI CDN — image, video, audio, or 3D file depending on the Seedream 5.5 variant you called.
Use cases
What you can build with Seedream 5.5
Common workflows developers and creators use the Seedream 5.5 API for.
Transparent assets straight from the model
Set background to transparent on bytedance/seedream-v5.5 or bytedance/seedream-v5.5/edit and the result is a PNG with an alpha channel, with no separate background-removal step. Edit can cut a subject out of an ordinary photo: the input image does not need an alpha channel.
Split an image into editable layers
bytedance/seedream-v5.5/layer-decomposition separates one image into a base image plus transparent PNG layers. max_images sets the total number of outputs, from 2 to 100 (30 by default), and an optional prompt tells the model which elements should become their own layers.
Multi-round editing that stays steady
Repeated edits on bytedance/seedream-v5.5/edit keep the areas you did not ask to change stable from one round to the next, so a product shot or portrait can be refined step by step without drifting.
Edits grounded in up to 10 references
Edit accepts 1 to 10 reference images per request: combine a product, a character, and a style reference in one call. With no aspect ratio set, the output follows the first input image.
Photorealistic images up to 3K
Seedream 5.5 is tuned for photographic realism and visual aesthetics. The 3k tier is about 8 megapixels (3840x2160 at 16:9), for hero images and print-sized deliverables; 1k and 1.5k keep drafts quick.
Small and multilingual text
Seedream 5.5 renders small text and text in multiple languages accurately: labels, packaging copy, signage, and localized posters that stay legible at output size.
Tips
Tips for prompting Seedream 5.5
Practical advice for getting better outputs from Seedream 5.5 — drawn from the patterns that work across image models in production pipelines.
- 01
Ask for transparency in the request, not the prompt
Set background to transparent and keep output_format at png: the model returns a PNG with an alpha channel. jpeg has no alpha channel, so a transparent request needs png. Leave background at auto when you do not mind either way.
- 02
Cut out subjects with Edit
bytedance/seedream-v5.5/edit with background set to transparent extracts the subject from an ordinary photo; the input does not need an alpha channel. Describe the subject to keep, such as "the sneaker, without the box", when the frame is busy.
- 03
Guide Layer Decomposition with a prompt
bytedance/seedream-v5.5/layer-decomposition works without a prompt, but naming the elements you want separated (headline text, product, logo area, background) gives you layers that match your design file. Lower max_images for a few clean layers; raise it for fine-grained separation.
- 04
Edit in rounds
Unedited areas stay steady across repeated edits, so make one change per request and feed the result back in. It is easier to review and keeps the rest of the image as it was.
- 05
Draft small, deliver at 3k
Settle the prompt at 1k or 1.5k, then re-run it at 3k (about 8 megapixels) for the final asset. Layer Decomposition tops out at 2k, so pick the decomposition resolution with that in mind.
- 06
Put exact text in quotes
Seedream 5.5 renders small and multilingual text accurately. Write the exact wording in quotes and in the target language, and say where it goes ("on the label", "top-left headline").
Pricing
Seedream 5.5 API pricing
Pricing is per-output. The final charge scales with the parameters you set in each variant's playground (resolution, duration, output count, references).
| Endpoint | Type | Starting price |
|---|---|---|
| bytedance/ | image-tools | $0.585 |
| bytedance/ | text-to-image | $0.0195 |
| bytedance/ | image-to-image | $0.027 |
API
Call the Seedream 5.5 API
Sign up for an API key at wavespeed.ai/accesskey, then submit a prediction via REST. The playground generates ready-to-paste samples for any combination of inputs.
POSThttps://api.wavespeed.ai/api/v3/bytedance/seedream-v5.5
# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
-X POST "https://api.wavespeed.ai/api/v3/bytedance/seedream-v5.5" \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $WAVESPEED_API_KEY" \
-d '{
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"aspect_ratio": "1:1",
"resolution": "1k",
"output_format": "png",
"background": "auto"
}')
TASK=$(printf '%s' "$SUBMIT_RESPONSE" | jq 'if has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "$TASK" | jq -r '.id')
if [ -z "$PREDICTION_ID" ] || [ "$PREDICTION_ID" = "null" ]; then
printf 'Submission response did not contain a prediction id
' >&2
exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/$PREDICTION_ID/result"
# 2. Poll until the prediction finishes.
while true; do
RESPONSE=$(curl --silent --show-error --fail-with-body "$RESULT_URL" \
-H "Authorization: Bearer $WAVESPEED_API_KEY")
RESULT=$(printf '%s' "$RESPONSE" | jq 'if has("data") then .data else . end')
STATUS=$(printf '%s' "$RESULT" | jq -r '.status')
case "$STATUS" in
completed) printf '%s\n' "$RESULT" | jq '.outputs'; break ;;
failed|cancelled|timeout|deleted) printf '%s\n' "$RESULT" | jq . >&2; exit 1 ;;
*) sleep 2 ;;
esac
doneconst submitUrl = "https://api.wavespeed.ai/api/v3/bytedance/seedream-v5.5";
const apiKey = process.env.WAVESPEED_API_KEY;
if (!apiKey) throw new Error('Set WAVESPEED_API_KEY');
async function requestJson(url, options = {}) {
const response = await fetch(url, options);
if (!response.ok) throw new Error(await response.text());
return response.json();
}
// 1. Submit the prediction.
const body = await requestJson(submitUrl, {
method: "POST",
headers: {
"Authorization": `Bearer ${apiKey}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"aspect_ratio": "1:1",
"resolution": "1k",
"output_format": "png",
"background": "auto"
}),
});
const task = body.data ?? body;
const resultUrl = `https://api.wavespeed.ai/api/v3/predictions/${task.id}/result`;
// 2. Poll until the prediction finishes.
while (true) {
const resultBody = await requestJson(resultUrl, {
headers: { "Authorization": `Bearer ${apiKey}` },
});
const result = resultBody.data ?? resultBody;
if (result.status === "completed") {
console.log(result.outputs);
break;
}
if (["failed", "cancelled", "timeout", "deleted"].includes(result.status)) throw new Error(JSON.stringify(result));
await new Promise(resolve => setTimeout(resolve, 2000));
}import json
import os
import time
from urllib.request import Request, urlopen
api_key = os.environ["WAVESPEED_API_KEY"]
headers = {"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"}
payload = {
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"aspect_ratio": "1:1",
"resolution": "1k",
"output_format": "png",
"background": "auto"
}
def request_json(url, data=None):
request = Request(url, data=data, headers=headers, method="POST" if data else "GET")
with urlopen(request) as response:
return json.load(response)
# 1. Submit the prediction.
body = request_json("https://api.wavespeed.ai/api/v3/bytedance/seedream-v5.5", json.dumps(payload).encode())
task = body.get("data", body)
result_url = f"https://api.wavespeed.ai/api/v3/predictions/{task['id']}/result"
# 2. Poll until the prediction finishes.
while True:
result_body = request_json(result_url)
result = result_body.get("data", result_body)
status = result.get("status")
if status == "completed":
print(result.get("outputs", []))
break
if status in {"failed", "cancelled", "timeout", "deleted"}:
raise RuntimeError(result)
time.sleep(2)Compare
Seedream 5.5 vs alternatives
When to pick Seedream 5.5 over similar models on WaveSpeedAI.
Seedream 5.5 vs Seedream 5 Pro
Seedream 5.5 adds three things 5 Pro does not have: native transparent output, a 3k resolution tier above 2k, and a layer decomposition endpoint that returns up to 99 transparent layers plus a base image. 5 Pro keeps a fast prompt-optimization mode that 5.5 does not offer.
Seedream 5.5 vs Nano Banana 2.1
Nano Banana 2.1 takes up to 14 reference images, reaches 4K, and offers optional web and image search. Seedream 5.5 answers with transparent PNG output from the model itself, a dedicated layer decomposition endpoint, and multi-round edits that keep unedited areas steady.
Seedream 5.5 vs GPT Image 2.6
GPT Image 2.6 has two tiers, five quality levels, animated GIF and vector SVG output, and edits with up to 16 references at up to 4K. Seedream 5.5 is the pick for transparent PNG assets, splitting finished images into layers, and photorealistic output up to 3K.
FAQ
Seedream 5.5 API — Frequently asked questions
Pricing, license, integration — common questions about running Seedream 5.5 on WaveSpeedAI.
Can Seedream 5.5 generate images with a transparent background?
Yes. Set background to transparent on text-to-image or Edit with png output, and the model returns a PNG with an alpha channel directly. For Edit, the input image does not need an alpha channel: the model can extract the subject from an ordinary photo.
What does Seedream 5.5 Layer Decomposition return?
A base image plus transparent PNG layers. max_images sets the total number of outputs, from 2 to 100 with 30 by default, so up to 99 layers. An optional prompt guides which elements become layers, and output_format applies to the base image; every layer is a PNG with transparency.
What is the difference between Seedream 5.5 and Seedream 5 Pro?
Seedream 5.5 adds native transparent output, a 3k resolution tier, and a layer decomposition endpoint with up to 99 layers. Seedream 5 Pro has a fast prompt-optimization mode that Seedream 5.5 does not offer.
What resolutions and formats does Seedream 5.5 support?
Text-to-image and Edit take 1k, 1.5k, 2k, and 3k (about 8 megapixels, such as 3840x2160 at 16:9); Layer Decomposition takes 1k, 1.5k, and 2k. Output is png (default) or jpeg.
What is the Seedream 5.5 API?
Seedream 5.5 is a ByteDance image generation model exposed as a REST API on WaveSpeedAI. ByteDance Seedream 5.5 — photographic realism and strong visual aesthetics, consistent multi-round editing that keeps unedited areas steady, and accurate small and multilingual text. Native transparent PNG output, a 3K resolution tier, and a layer decomposition endpoint that splits one image into a base image plus transparent layers. You can call it programmatically or try it from the playground linked above.
How do I call the Seedream 5.5 API?
Sign up for a WaveSpeedAI account, copy your API key from /accesskey, then POST to https://api.wavespeed.ai/api/v3/bytedance/seedream-v5.5 with your input as JSON. The endpoint returns a prediction id. Poll the result endpoint starting around every 2 seconds, increase the interval for long-running tasks, and stop on any terminal status. Production-oriented Python / Node.js / cURL examples are above.
How much does the Seedream 5.5 API cost?
Seedream 5.5 starts at $0.0195. The exact cost scales with the parameters you set (resolution, duration, output count, references). The live cost preview next to the Generate button in the playground shows the exact price for your current input.
Which Seedream 5.5 variants are available?
WaveSpeedAI hosts 3 live Seedream 5.5 endpoints: bytedance/seedream-v5.5/layer-decomposition, bytedance/seedream-v5.5, bytedance/seedream-v5.5/edit. Each variant has its own playground page and pricing.
Can I use Seedream 5.5 outputs commercially?
Commercial usage rights follow the ByteDance model license. Most ByteDance models permit commercial output use; see each model's playground page for the specific license summary, and WaveSpeedAI's Terms of Service for platform-level conditions.
Why use Seedream 5.5 on WaveSpeedAI instead of going direct?
One API key + one billing account across Seedream 5.5 AND 1,000+ other AI models from other providers. No per-vendor SDK setup, no separate rate-limit envelopes, no rewrite-per-vendor integration code. Pricing is typically at parity with or below ByteDance's direct API.
Provider
About ByteDance
The team behind Seedream 5.5 and the broader ByteDance model lineup on WaveSpeedAI.
ByteDance — the company behind TikTok, CapCut, and Doubao — runs one of the largest AI research operations in the world. Its Seed and Doubao teams ship state-of-the-art models for video generation (Seedance), image generation and editing (Seedream, Nano Banana), and avatar / lip-sync work. Models are trained on internal-scale data and shipped to production through ByteDance's consumer apps before being released as APIs.
Start building with Seedream 5.5 on WaveSpeedAI
Free starter credits on signup. One API key across 1,000+ AI models from ByteDance and every other provider.
/filters:quality(82)/examples/52d1777161efd4c4a1eb8f367d2bfdc1/e4e62e6c-afed-4fc6-81e3-8e15d9a2905c-a90fc868dcee.webp)
/filters:quality(82)/examples/a3a2c988c8ff271c6311fc6fd5b75076/8c70bf02-7370-4785-a80a-247fe5b0cdf5-487bb3f09732.webp)
/filters:quality(82)/examples/68f10dbf5606d977f7c60a8c6cc321db/fc56bf35-bdd2-481b-8650-251ff7a3a4ff-1ffd75fef9ed.webp)
/filters:quality(82)/examples/5549abed42d94882c754d980220eed45/8cc56cee-6f83-44cd-8533-7f921932a352-0c5fbb46e60d.webp)