Seedance 2.5 API. Clips from $0.18.

One clip on each model

The $2 trial credit makes about 11 Seedance 2.0 Fast clips.

Seedance API prices per second

Every model is charged by the second of video it makes. BytePlus's list is converted to the same clip size.

What the API makes today

Each model makes one clip size from a text prompt. The size comes with the model, so there is nothing to set.

• Text to video: One text prompt per video. Image and video references are not supported.
• Landscape clips: Seedance 2.5 and Seedance 2.0 Fast: 4 seconds, 480p, 16:9. For drafts, storyboards and B-roll.
• Vertical clips: Seedance 2.0 Mini: 5 seconds, 720p, 9:16. Sized for Shorts, Reels and TikTok.
• Charged when it works: A video that fails is not charged: the credit it reserved goes back to your balance.

Make a video in three calls

• Get a key: Start free on the $2 trial credit, or top up from $10 below. A top-up key arrives by email.
• Create a task: POST /v1/videos/generations with a model and a prompt. It answers 202 with a task_id.
• Poll and download: GET /v1/videos/tasks/{task_id} every 10 seconds or so. When the status is completed, the video is at result.url.

import time, requests

API = "https://api.apisrouter.com/v1"
headers = {"Authorization": "Bearer sk-your-apisrouter-key"}

task = requests.post(f"{API}/videos/generations", headers=headers, json={
    "model": "seedance-2.5",
    "prompt": "A red fox runs through fresh snow at dawn, low tracking shot",
}).json()

while task["status"] == "processing":
    time.sleep(10)
    task = requests.get(f"{API}/videos/tasks/{task['task_id']}", headers=headers).json()

print(task["result"]["url"] if task["status"] == "completed" else task["error"])

Two ways to pay

Both fill one balance, spent per second of video and per token on text models.

FAQ