Seedance Video 2.5 API

Seedance Video 2.5 is a model from MiniMax that you can call through the ScalingTensor API.

Provider

MiniMax

Billing

Provider cost

ScalingTensor price

Provider cost USD / 1M tokens; varies by resolution and video input

Playground

Try Seedance Video 2.5

Edit the params, add your API key and run the model here. The result plays on this page when the job completes.

Pricing

Seedance Video 2.5 API pricing

Prices come from the live catalog. There is no subscription; each request is charged to your prepaid wallet.

ScalingTensor price

Provider cost

Live catalog price

Provider cost

USD / 1M tokens; varies by resolution and video input

Compare all model prices

Shared balance

Add credits

Top up

One wallet and one API key for every model in the catalog. Top-ups start at $10 and are paid through Stripe.

$10 $20 $50 Custom

Parameters

Input parameters

These parameters belong inside the request body's params object.

Input parameters
name	type	required	description
`model`	`string`	true	Public model name. Must be nova-video-2.5.
`content`	`array<object>`	true	Multimodal input array. Supports text, image, video, and audio, including audio-only input.
`content[].type`	`string`	true	Content item type: text, image_url, video_url, or audio_url.
`content[].text`	`string`	conditional	Prompt text when type is text.
`content[].image_url.url`	`string`	conditional	Image URL or inline data:image/...;base64,... when type is image_url.
`content[].video_url.url`	`string`	conditional	Video URL when type is video_url. Including a video input selects video-to-video billing.
`content[].audio_url.url`	`string`	conditional	Audio URL when type is audio_url; nova-video-2.5 supports audio-only input.
`content[].role`	`string`	false	Optional role tag, such as first_frame, reference_image, reference_video, or reference_audio.
`callback_url`	`string`	false	Callback URL for queued, running, succeeded, failed, or expired status changes.
`return_last_frame`	`boolean`	false	Whether to return the generated video's last frame. Defaults to false.
`service_tier`	`string`	false	Service tier. Defaults to default.
`execution_expires_after`	`integer`	false	Task timeout in seconds, from 3600 to 259200. Defaults to 172800.
`generate_audio`	`boolean`	false	Whether to generate synchronized audio. Defaults to true.
`safety_identifier`	`string`	false	Hashed end-user identifier; maximum length 64 characters.
`resolution`	`string`	false	Output resolution: 480p or 720p. Defaults to 720p; 1080p is unsupported.
`ratio`	`string`	false	Aspect ratio: 16:9, 9:16, 4:3, 3:4, 1:1, 21:9, or adaptive. Defaults to adaptive. First-frame and video editing tasks only support adaptive; video editing and extension require duration -1.
`duration`	`integer`	false	Video duration in seconds, from 4 to 30, or -1 for model-selected duration. Defaults to 5.
`seed`	`integer`	false	Random seed from -1 to 2^32-1. Defaults to -1 for a random seed.
`watermark`	`boolean`	false	Whether to add an AI-generated watermark. Defaults to false.
`tools`	`array<object>`	false	Optional tools array, for example [{"type":"web_search"}]. Defaults to an empty array.
`output_format`	`string`	false	Output format: mp4 or mov. Supported only by nova-video-2.5.
`task_id`	`string`	true for query	Task ID in the query endpoint path, returned by the create endpoint.

Output fields

Output fields
name	type	description
`id`	`string`	Task ID returned by task creation and task queries.
`model`	`string`	Model name used to create the task.
`status`	`string`	Task status: queued, running, in_progress, succeeded, completed, failed, or expired.
`content.video_url`	`string`	Generated video URL when successful; valid for 24 hours.
`content.last_frame_url`	`string`	Generated last-frame image URL when requested and available.
`usage.completion_tokens`	`integer`	Billed completion token count.
`usage.total_tokens`	`integer`	Total token count; currently equal to completion_tokens.
`error.code`	`string`	Error code when the task fails.
`error.message`	`string`	Error details when the task fails.
`created_at`	`integer`	Task creation time as a Unix timestamp in seconds.
`updated_at`	`integer`	Most recent task update time as a Unix timestamp in seconds.
`resolution`	`string`	Actual output resolution.
`ratio`	`string`	Actual output aspect ratio.
`duration`	`integer`	Actual output duration in seconds.
`seed`	`string`	Actual generation seed.
`generate_audio`	`boolean`	Whether synchronized audio was generated.
`callback_url`	`string`	Callback URL echoed in the task response, when supplied.

API

Call Seedance Video 2.5 from your code

Create a key in the console and call the API from your server using the apiKey header. Save data.jobId and poll GET /api/media/result?jobId=... with the same key; a code of 0 means success.

POST /api/media/generate

curl --request POST 'https://restapi.scalingtensor.com/api/media/generate' \
  --header 'apiKey: YOUR_API_KEY' \
  --header 'Content-Type: application/json' \
  --data-raw '{
  "model": "nova-video-2.5",
  "params": {
    "model": "nova-video-2.5",
    "ratio": "16:9",
    "content": [
      {
        "text": "A beach at sunset",
        "type": "text"
      }
    ],
    "duration": 5,
    "resolution": "720p",
    "output_format": "mp4",
    "generate_audio": true
  }
}'

const response = await fetch("https://restapi.scalingtensor.com/api/media/generate", {
  method: "POST",
  headers: {
    apiKey: process.env.SCALINGTENSOR_API_KEY,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "model": "nova-video-2.5",
    "params": {
      "model": "nova-video-2.5",
      "ratio": "16:9",
      "content": [
        {
          "text": "A beach at sunset",
          "type": "text"
        }
      ],
      "duration": 5,
      "resolution": "720p",
      "output_format": "mp4",
      "generate_audio": true
    }
  }),
});

const result = await response.json();
if (!response.ok || result.code !== 0) {
  throw new Error(result.msg || `Request failed: ${response.status}`);
}
console.log(result.data); // Save data.jobId to query /api/media/result

import os
import requests

payload = {
    "model": "nova-video-2.5",
    "params": {
        "model": "nova-video-2.5",
        "ratio": "16:9",
        "content": [
            {
                "text": "A beach at sunset",
                "type": "text",
            },
        ],
        "duration": 5,
        "resolution": "720p",
        "output_format": "mp4",
        "generate_audio": True,
    },
}

response = requests.post(
    "https://restapi.scalingtensor.com/api/media/generate",
    headers={"apiKey": os.environ["SCALINGTENSOR_API_KEY"]},
    json=payload,
    timeout=60,
)
result = response.json()
if result["code"] != 0:
    raise RuntimeError(result["msg"])
print(result["data"])  # Save data["jobId"] to query /api/media/result

API documentation · This page as Markdown · Markdown API reference

Related models

Seedance Video 2.5 API FAQ

How much does the Seedance Video 2.5 API cost?

Seedance Video 2.5 is billed at the provider's actual cost for each job (USD / 1M tokens; varies by resolution and video input). The charge is deducted from your prepaid ScalingTensor wallet.

How do I call the Seedance Video 2.5 API?

Send POST /api/media/generate with "model": "nova-video-2.5" and the parameters in a params object, using your apiKey header. The call returns a jobId. Poll GET /api/media/result?jobId=... until the status is completed, then read the output URL from the result.

Which parameters does Seedance Video 2.5 accept?

Required: model and content. Optional: callback_url, return_last_frame, service_tier, execution_expires_after, generate_audio, safety_identifier, resolution, ratio, duration and seed and 4 more. The parameter table on this page describes each one.

Do I need a MiniMax account to use Seedance Video 2.5?

No. You call Seedance Video 2.5 with your ScalingTensor API key and pay from your ScalingTensor wallet. The same key works for every other model in the catalog.