Seedream 4.0 Prompts — 8 Tested Recipes With Outputs

from $0.0195 per image — one OpenAI-compatible endpoint at https://api.apimart.ai/v1, no monthly plan required. (observed 2026-10-12)

Get an API key · Live pricing · Model page · ⚡ 60-second quickstart

Why teams call Seedream 4.0 (seedream-4-0) through APIMart

• One key, entire catalog. The same https://api.apimart.ai/v1 base URL and Authorization header reach Seedream 4.0 (seedream-4-0) and 300+ other image, video and language models — switch the model field, not your client.
• $1 minimum, pay as you go. No subscription and no prepaid plan to size up front: top up from $1 and spend it on calls.
• The charge comes back in the response. Every completed task reports the amount billed (cost / credits_cost); a task marked failed is refunded.
• Async by design. Submit, take the task_id, poll GET /v1/tasks/{id} — batching and retries are ordinary queue work, not a bespoke integration.

Seedream 4.0 is ByteDance's unified text-to-image and image editing model — 4K output, industry-leading text rendering, batch processing and multi-reference support.

Model id and routes

Route	model value	Billing	Notes
Per-image	seedream-4-0	per delivered image, by resolution	see the model page for the exact tiers

Endpoint: POST https://api.apimart.ai/v1/images/generations (OpenAI-compatible), then poll GET /v1/tasks/{task_id}.

Pricing

Output	List price	Effective price
default	0.024375	0.0195

Prices are a snapshot; the pricing page and data/model.json are refreshed by CI, and a completed task reports the exact amount in its cost field.

Request parameters

Field	Type	Default	Notes
model	string	required	the model id shown in the route table above
prompt	string	required	what to generate, or what to change when editing
size	string	auto / 1:1	aspect ratio, or auto where the route supports it
resolution	string	1K	output resolution tier — cost scales with it
n	integer	1	number of images per task, where the route allows more than one
image_urls	string[]	—	reference images as public HTTP(S) URLs or base64 data URLs
nsfw_check	boolean	false	runs omni-moderation-latest before submitting

The table lists the fields accepted by the APIMart OpenAI-compatible image endpoint. Check the model page for anything this model adds on top.

Quickstart

curl --request POST --url https://api.apimart.ai/v1/images/generations \
  --header "Authorization: Bearer $APIMART_API_KEY" --header 'Content-Type: application/json' \
  -d '{"model":"seedream-4-0","prompt":"A bamboo forest at sunrise, slow push-in","size":"1:1","resolution":"1K"}'

import os, time, requests

BASE = "https://api.apimart.ai/v1"
HEADERS = {"Authorization": f"Bearer {os.environ['APIMART_API_KEY']}", "Content-Type": "application/json"}

created = requests.post(f"{BASE}/images/generations", headers=HEADERS, timeout=60, json={
    "model": "seedream-4-0", "prompt": "A bamboo forest at sunrise, slow push-in",
}).json()
task_id = created["data"]["id"]

while True:
    task = requests.get(f"{BASE}/tasks/{task_id}", headers=HEADERS, timeout=60).json()["data"]
    if task["status"] in ("completed", "failed"):
        break
    time.sleep(5)
print(task.get("cost"), task["result"]["images"][0]["url"])

Runnable versions: examples/. The task lifecycle is pending → processing → completed | failed.

FAQ

What is the Seedream 4.0 model id?

seedream-4-0. See the route table above.

How is it billed?

from $0.0195 per image. The completed task reports the exact amount in its cost field; a failed task is refunded.

How do I get a key?

Use the Get an API key link — $1 minimum top-up, no plan required.

Attributed links (how this repository is measured)

Purpose	Attributed link	Target
Model page	https://go.apimart.ai/k-4fd790	apimart.ai/zh/model/seedream-4
Live pricing	https://go.apimart.ai/k-a71e73	apimart.ai/zh/pricing
Get an API key	https://go.apimart.ai/k-2e8379	apimart.ai/zh/keys

Outbound APIMart links are minted through the promo link API; hand-made tracking parameters are rejected by tools/check_links.py in CI.

Disclosure

Seedream 4.0 is a third-party model served through APIMart; this repository documents how to call it and publishes model ids and prices, and does not claim official status. Model names, prices and documentation belong to their respective owners.

License

MIT — see LICENSE.