ByteDance Seed: Seedream 5.0 Flash
Seedream 5.0 Flash creates and edits images from text and visual references, with spatially guided changes, multilingual image text and layer decomposition. ByteDance Seed positions Flash for faster generation, while keeping it distinct from the quality-focused Pro edition.
About the model
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
A faster image-creation edition
Seedream 5.0 Flash is ByteDance Seed’s image-generation and editing model for workflows that prioritize generation speed. The maker’s Pro/Flash guide describes both text-to-image and single- or multi-reference image-to-image creation. It distinguishes Flash’s faster generation from Pro’s higher image-quality priority, without establishing a fixed response time for every brief.
The defining shared features are interactive editing, layer decomposition and native multilingual generation. Ordinary generation returns one finished image. The manufacturer’s Flash workflow does not support sequential related-image generation, web search or streaming image output. Its prompt optimization uses standard mode; the separate fast optimization control belongs to Pro rather than Flash.
Creating from a brief and references
State the intended picture, the main subject, composition and visual treatment, then identify what each reference controls. For a product visual, separate the reference that establishes the object from one that supplies a palette or surface material. “Keep the object’s silhouette and viewing angle from the first image; use the second image only for the muted background palette” gives the references distinct roles.
The maker recommends concise prompts, no more than 300 Chinese characters or 600 English words. That is preparation guidance rather than a stated hard text-input cutoff. List the decisive requirements first, including exact wording that must appear, the desired aspect ratio and details that must stay unchanged. Inspect the finished image against that list rather than rely only on its overall appearance.
Spatially guided editing
Flash can use points, bounding boxes, arrows and free-form marks to identify where an edit should happen. Supply the source image together with a location and the change. For an interior concept, mark the chair, request a fabric replacement and identify the floor, wall, camera angle and nearby objects to preserve. If the marks are guidance only, explicitly ask for their removal from the finished image.
The maker’s coordinate-editing guide uses <point> and <bbox> tags with coordinates normalized to 0–999 relative to the specified input image. A point contains x and y; a box contains the upper-left and lower-right coordinates. Identify the image number and target object, especially when a box includes several subjects. For a cross-image edit, distinguish the object to take from one reference and the destination region in another.
Use the displayed image’s own dimensions when obtaining the proportional location; canvas zoom, panning or browser coordinates are not the normalized image coordinates. Describe protected regions as well as the edited region. Compare the result’s perspective, illumination and surrounding detail with the original source.
Separate a composition into editable layers
Layer decomposition is a distinct mode that takes exactly one image and returns a base image plus up to 16 independent layers. Each layer is a PNG with alpha. A prompt can name the elements to separate, such as title, main subject and decorations; without a decomposition prompt, the model identifies the main elements automatically.
The results include layer names, descriptions, stacking order and placement boxes. Use the base image with stacking index zero, then place the other layers in ascending stacking order. Absolute boxes describe placement in the output base image; normalized boxes describe proportional placement on a different canvas. Keep this returned placement metadata with the image files so the original composition can be reconstructed before it is rearranged.
For a poster revision, separate the main artwork from the title and decorations, then inspect edges and any restored background area. A transparent layer can be edited as a single alpha-channel input with transparent PNG output. The transparent-background option is scoped to that single-input image-editing workflow; ordinary text-to-image generation is a separate operation.
Multilingual composition
In addition to Chinese and English, the maker documents native prompt and image-text support for Russian, Arabic, Filipino, Thai, Turkish, Korean, Malay, Spanish, Portuguese, Indonesian, French, German, Vietnamese and Japanese. Supply the exact approved wording and reading direction, and identify which region contains each text block. Review accents, joined characters, line breaks and small labels in the final image.
Reference files and output specifications
Image generation accepts up to 10 references in JPEG, PNG, WebP, BMP, TIFF, GIF, HEIC or HEIF. Each can be up to 30 MB, with both sides above 14 pixels, 196–36,000,000 total pixels and an aspect ratio from 1:16 to 16:1. Layer decomposition instead requires one JPEG or PNG of 262,144–36,000,000 pixels, up to 30 MB, within the same ratio range.
Generation offers 1K, 1.5K and 2K output tiers, defaulting to 2K, or explicit dimensions with 921,600–4,624,220 total pixels and ratios from 1:16 to 16:1. Both constraints apply to custom dimensions. Layer output uses 1K, 1.5K, 2K or auto, with auto as the default; the base and each layer retain their respective source aspect ratios.
Finished images are JPEG or PNG. The base image’s format can be selected separately, while decomposed layers remain PNG with alpha. Delivery can use encoded image data or download links retained for 24 hours. Save every returned image and its metadata for the intended design workflow. Inspect text, edited details and layer alignment at the size where the artwork will be used.
Ailoot Ultra · Images POST /v1/images/generations bytedance-seed/seedream-5-0-flash
bytedance-seed/seedream-5-0-flash
bytedance-seed/seedream-5-0-flashOverview
Create one finished image from a text brief.
- Create a product still
- Explore a visual concept
- Illustrate a scene
- Accepted inputs
- Text
- Results
- Images
Available in Ailoot · inputs & methods
Ailoot Ultra · Images POST /v1/images/generations
- Input types
text- Output types
image- Output format
- Image file
- Output count
- 1 output
| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits bytedance-seed/seedream-5-0-flash | Default— |
Fieldprompt | Typestring | RequirementRequired | Values & limits Minimum length: 1 Conditions: Nonempty UTF8 prompt. No manufacturer-wide character limit is invented. | Default— |
Fieldn | Typeinteger | RequirementOptional | Values & limits 1 | Default1 |
Fieldresponse_format | Typestring | RequirementOptional | Values & limits url, b64_json | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits false | Defaultfalse |
Input material combinations
Text to image · Required: prompt
Method parameters and lifecycle
| Parameter | Contract |
|---|---|
Parametercontent_type | Contractapplication/json |
Parameterrequired | Contractmodel, prompt |
Parametermaterial_types | Contracttext |
Parameteroptional | Contractn, response_format, stream |
Parametern | Contract1 |
Parameterstream | Contractfalse |
Parameterresponse_format | Contracturl, b64_json |
Parameterreference_images | Contractfalse |
Parameteredit | Contractfalse |
Parameteroutput_format_optional | Contract |
Parameterresolution_control | Contractnull |
Parameteraspect_ratio_control | Contractnull |
Parameterother_fields | Contractrejected except exact pinned internal provider routing |
Lifecycle: synchronous_result_then_owned_library
GET /v1/media/tasks/{task_id} · actual returned task_id or media library record
GET /v1/media/assets/{asset_id}/content
GET /v1/media/assets
GET /v1/media/assets/{asset_id}
Generate one image from prompt.
Save every data item; url/b64_json are result encodings.
Pricing
How the final charge is calculated
The displayed rates estimate the cost before generation. The final charge is calculated from the cost recorded for the original generation. Measured output duration and forecast quantities are not the final billable quantity.
Tariff references do not add request options. Supported parameters are listed for the selected method.
- Fast mode
- Not specified When fast mode is selected
- Flex mode
- Not specified When flex mode is selected
- Maximum reasoning effort
- Not specified When maximum reasoning effort uses the default adjustment
Estimated rates for comparison
Prices by configuration
Match each charge to the conditions shown. Use one rate for each set of alternative configurations.
| Configuration | Rate schedule | Forecast quantity | Billing unit | Charge |
|---|---|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Forecast quantity— | Billing unit— | Charge Input · Input tokens · Estimate for comparison only — / 1K tokens Output · Output tokens · Estimate for comparison only — / 1K tokens Input · Cache read · Estimate for comparison only — / 1K tokens Input · Cache write · Estimate for comparison only — / 1K tokens Input · Cache write, 1 hour · Estimate for comparison only — / 1K tokens |
| Configurationinput image | Rate schedule— | Forecast quantity— | Billing unitper input image | Charge Input · input image · Estimate for comparison only $0 per input image |
| ConfigurationUltra | Rate schedule— | Forecast quantity— | Billing unitper image output | Charge Output · image output · Estimate for comparison only $0.036 per image output |
| ConfigurationEstimated generation cost | Rate schedule— | Forecast quantityone generation estimate | Billing unit— | Charge Output · Estimated generation cost · Estimate for comparison only $0.036 estimate before generation This estimate is not an additional charge or a fixed rate for the completed generation. |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
Reference tariff details
Model-level reference schedules do not change method selection, quotes or charges. Active charges use the selected Ailoot method rates.
Image pricing configurations
Prices by configuration
Match each charge to the conditions shown. Use one rate for each set of alternative configurations.
| Configuration | Billing conditions | Charge |
|---|---|---|
| ConfigurationOutput image | Billing conditionsNo pricing variant is declared. | Charge Output · Output image $0.036 per image |
| ConfigurationInput image | Billing conditionsNo pricing variant is declared. | Charge Input · Input image $0 per image |
Equivalent listings describe the same charge; they do not add a second fee.
Example requests
Ailoot Ultra · Images
set -eu
: "${AILOOT_API_KEY:?Set your Ailoot API key}"
export AILOOT_API_KEY
python3 - <<'PY_REQUEST'
import json,os
from pathlib import Path
p=json.loads('{"model": "bytedance-seed/seedream-5-0-flash", "prompt": "A lighthouse on a cliff, watercolor", "n": 1, "response_format": "b64_json"}')
Path("request.json").write_text(json.dumps(p,ensure_ascii=False))
PY_REQUEST
curl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/images/generations \
-H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \
--data-binary @request.json -D response-headers.txt -o result.json
python3 - <<'PY_RESULT'
import base64, json, os, subprocess
from pathlib import Path
from urllib.parse import urlsplit, urljoin
def save_images(items):
if not items: raise SystemExit("No completed images; retain the full response")
for index, item in enumerate(items):
encoded=item.get("b64_json")
uri=item.get("url", "")
if not encoded and uri.startswith("data:image/"):
header, separator, encoded=uri.partition(",")
if not separator or ";base64" not in header: raise SystemExit("Unsupported image data URI")
if encoded:
raw=base64.b64decode(encoded,validate=True)
elif uri:
uri=urljoin("https://api.ailoot.ai",uri)
parsed=urlsplit(uri)
if parsed.scheme!="https" or not parsed.netloc or parsed.username or parsed.password:
raise SystemExit("Invalid returned image URL")
owned=parsed.netloc=="api.ailoot.ai" and parsed.path.startswith("/v1/media/assets/") and parsed.path.endswith("/content")
args=["curl","--fail-with-body","--silent","--show-error",uri]
if owned: args += ["-H","Authorization: Bearer "+os.environ["AILOOT_API_KEY"]]
temporary=Path(f"image-{index}.bin")
subprocess.run(args+["-o",str(temporary)],check=True)
raw=temporary.read_bytes()
temporary.unlink()
else: raise SystemExit("Image has neither bytes nor URL")
if not raw: raise SystemExit("Empty image result")
ext="png" if raw.startswith(b"\x89PNG") else "jpg" if raw.startswith(b"\xff\xd8") else "webp" if raw[:4]==b"RIFF" and raw[8:12]==b"WEBP" else "svg" if raw.lstrip().startswith((b"<svg",b"<?xml")) else "bin"
Path(f"image-{index}.{ext}").write_bytes(raw)
print("Every returned image saved; complete response metadata retained")
save_images(json.loads(Path("result.json").read_text())["data"])
PY_RESULT
from openai import OpenAI
import base64, json, os, subprocess
from pathlib import Path
from urllib.parse import urlsplit, urljoin
def save_images(items):
if not items: raise SystemExit("No completed images; retain the full response")
for index, item in enumerate(items):
encoded=item.get("b64_json")
uri=item.get("url", "")
if not encoded and uri.startswith("data:image/"):
header, separator, encoded=uri.partition(",")
if not separator or ";base64" not in header: raise SystemExit("Unsupported image data URI")
if encoded:
raw=base64.b64decode(encoded,validate=True)
elif uri:
uri=urljoin("https://api.ailoot.ai",uri)
parsed=urlsplit(uri)
if parsed.scheme!="https" or not parsed.netloc or parsed.username or parsed.password:
raise SystemExit("Invalid returned image URL")
owned=parsed.netloc=="api.ailoot.ai" and parsed.path.startswith("/v1/media/assets/") and parsed.path.endswith("/content")
args=["curl","--fail-with-body","--silent","--show-error",uri]
if owned: args += ["-H","Authorization: Bearer "+os.environ["AILOOT_API_KEY"]]
temporary=Path(f"image-{index}.bin")
subprocess.run(args+["-o",str(temporary)],check=True)
raw=temporary.read_bytes()
temporary.unlink()
else: raise SystemExit("Image has neither bytes nor URL")
if not raw: raise SystemExit("Empty image result")
ext="png" if raw.startswith(b"\x89PNG") else "jpg" if raw.startswith(b"\xff\xd8") else "webp" if raw[:4]==b"RIFF" and raw[8:12]==b"WEBP" else "svg" if raw.lstrip().startswith((b"<svg",b"<?xml")) else "bin"
Path(f"image-{index}.{ext}").write_bytes(raw)
print("Every returned image saved; complete response metadata retained")
request={'model': 'bytedance-seed/seedream-5-0-flash', 'prompt': 'A lighthouse on a cliff, watercolor', 'n': 1, 'response_format': 'b64_json'}
client=OpenAI(base_url="https://api.ailoot.ai/v1",api_key=os.environ["AILOOT_API_KEY"],max_retries=0)
result=client.images.generate(**request)
Path("sdk-result.txt").write_text(str(result),encoding="utf-8")
save_images([{"b64_json":getattr(item,"b64_json",None),"url":getattr(item,"url",None)} for item in result.data])
import OpenAI from "openai";
import { readFileSync, writeFileSync } from "node:fs";
const key=process.env.AILOOT_API_KEY;
if(!key) throw new Error("Set your actual Ailoot API key");
const client=new OpenAI({baseURL:"https://api.ailoot.ai/v1",apiKey:key,maxRetries:0});
const request={"model": "bytedance-seed/seedream-5-0-flash", "prompt": "A lighthouse on a cliff, watercolor", "n": 1, "response_format": "b64_json"};
const result=await client.images.generate(request);
writeFileSync("sdk-result.json",JSON.stringify(result,null,2));
if(!result.data?.length) throw new Error("No completed images");
for(let index=0;index<result.data.length;index++) {
const item=result.data[index];let encoded=item.b64_json,uri=item.url||"",raw;
if(!encoded && uri.startsWith("data:image/")) {
const comma=uri.indexOf(",");if(comma<0 || !uri.slice(0,comma).includes(";base64")) throw new Error("Unsupported image data URI");
encoded=uri.slice(comma+1);
}
if(encoded) {
if(encoded.length%4!==0 || !/^[A-Za-z0-9+/]*={0,2}$/.test(encoded)) throw new Error("Invalid Base64");
raw=Buffer.from(encoded,"base64");
} else if(uri) {
const u=new URL(uri,"https://api.ailoot.ai");
if(u.protocol!=="https:" || u.username || u.password) throw new Error("Invalid returned URL");
const owned=u.origin==="https://api.ailoot.ai" && u.pathname.startsWith("/v1/media/assets/") && u.pathname.endsWith("/content");
const response=await fetch(u,{headers:owned?{Authorization:"Bearer "+key}:{},redirect:"manual"});
if(!response.ok) throw new Error("Image content download failed: "+response.status);
raw=Buffer.from(await response.arrayBuffer());
} else throw new Error("Image has neither bytes nor URL");
if(!raw.length) throw new Error("Empty image result");
const ext=raw.subarray(0,4).equals(Buffer.from([137,80,78,71]))?"png":raw[0]===255&&raw[1]===216?"jpg":raw.toString("ascii",0,4)==="RIFF"&&raw.toString("ascii",8,12)==="WEBP"?"webp":raw.toString("utf8").trimStart().startsWith("<svg")||raw.toString("utf8").trimStart().startsWith("<?xml")?"svg":"bin";
writeFileSync(`image-${index}.${ext}`,raw);
}
console.log("Every returned image saved; sdk-result.json retains complete metadata");
Receive and save results
Ailoot Ultra · Images
Download each returned image or save its Base64 contents
GET /v1/media/tasks/{task_id}GET /v1/media/assets/{asset_id}/contentGET /v1/media/assetsGET /v1/media/assets/{asset_id}
Keep all completed results
Save every completed image; retrieve each returned URL or decode its Base64 contents.
Save every data item. Decode each b64_json or fetch every returned signed URL. Select a filename extension from the actual MIME/bytes.
If the original response needs checking
If an image generation or edit result requires a check, keep the original operation_id/X-Client-Request-ID and returned result information. Recover existing owned results through the media library using actual task_id or asset_id values. Do not resubmit the generation/edit request or construct a polling URL from operation_id.
Response fields
| Field | Type |
|---|---|
Fieldcreated | Typeinteger |
Fielddata[].url | Typestring |
Fielddata[].b64_json | Typebase64 string |
Fieldusage | Typeobject |
Errors and next steps
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionInvalid required fields, material representation or parameter value | Next stepCorrect the indicated fields before a new creation request. |
Responseserver_error | ConditionThe returned response says the operation or result requires a check | Next stepKeep the original returned operation/task ID, request headers and response; continue supported reads of that original operation. Do not resubmit the creation request. |
Aliases
Model aliases
Ailoot Ultra bytedance-seed/seedream-5-0-flash
Method path aliases
Use the method paths shown above.
Related caller IDs
FAQ · ByteDance Seed: Seedream 5.0 Flash
- What inputs and results are available?
POST /v1/images/generations·bytedance-seed/seedream-5-0-flash:text→image- Which fields does the request use?
POST /v1/images/generations·bytedance-seed/seedream-5-0-flash:model,prompt,n,response_format,stream