ByteDance logo

Seedance 2.5 API

seedance-2.5

Seedance 2.5 API is APINEED's newer AITokenDance video route for 4–15 second text, image, and video-reference generation at 480p or 720p.

The APINEED production adapter deliberately exposes a narrower contract than third-party descriptions of the broader model: 480p and 720p output, up to 15 seconds, explicit metadata controls, optional reference media, and asynchronous polling. Unsupported 1080p requests are rejected before they can be billed.

MODALITIES
PriceUsage-based pricing · final cost depends on request and usageCONTEXTN/A

Playground

Generate with Seedance 2.5, review the output, and use the same request shape in your application.

Generation controls are waiting for verified model capability metadata.
Workflow
Usage-based pricing · final cost depends on request and usageSign in required. Final cost uses completed usage.
Output Ready
Seedance family sample
Request preview
{
  "model": "seedance-2.5",
  "workflow": "text_to_video",
  "input": {
    "prompt": "A cinematic product shot with deliberate camera movement, natural lighting, and a clear beginning and end."
  },
  "output": {
    "duration_seconds": 1
  }
}

Providers

ByteDance documents the broader model capability. APINEED separately defines the model ID, endpoint, parameter whitelist, authentication, and billing behavior shown on this page.

API NeedUsage-based
Uptime
Total Context
N/A
Max Output
N/A
Input
Usage-based/ 1M
Output
Usage-based/ 1M
Cache Read
Route priced
Cache Write
Route priced

Discount

Compare the published baseline with APINEED pricing for Seedance 2.5 API. Final generation cost follows completed server usage. [1][3]

Availability

APINEED monitors Seedance 2.5 API availability. Applications should preserve request and task IDs and handle upstream failures explicitly.

System statusLast 90 days
APINEED GATEWAY99.91% uptime

Seedance 2.5 API capabilities

Seedance 2.5 is presented here through the controls APINEED has actually validated in production, keeping the public playground aligned with the gateway rather than speculative provider features. [1][2][3]

Bounded production settings

APINEED accepts integer-equivalent duration from four through fifteen seconds and limits output resolution to 480p or 720p, rejecting unsupported combinations before task submission. [1][3]

Text and reference workflows

The route supports prompt-only generation, reference images, and one structured video reference while preserving a single task shape across the visible workflow choices. [1][3]

Canonical billing input

Duration, resolution, and video-reference presence are normalized before pre-consumption so the same request properties drive validation, upstream generation, and final pricing. [3]

Asynchronous completion

A successful create call returns a task ID rather than the finished clip. Applications poll the generation endpoint and read the nested video URL after the task reaches success. [1][3]

Condition-aware 3% discount

The production expression prices video-input and no-video-input work separately, then applies the configured 97% rate to the provider baseline for the supported resolutions. [1][3]

Building reliable Seedance 2.5 jobs

Stay within the verified ceiling. Use 480p or 720p and a duration from four to fifteen seconds; do not send 1080p as a hopeful fallback.

Plan one concise sequence. Describe the opening state, motion, camera behavior, sound, and final state in chronological order.

Persist task context. Store the task ID with the originating model and request so polling can resume safely after a page reload.

Quick Start

Call Seedance 2.5 with the verified APINEED endpoint and model-specific request shape.

1

Create an API key

Store your APINEED key server-side as API_NEED_API_KEY.

export API_NEED_API_KEY=sk-apineed-v1-...
2

Send the request

Use seedance-2.5 with POST /v1/media/generations.

curl https://apineed.com/v1/media/generations \
  -H "Authorization: Bearer $API_NEED_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "seedance-2.5",
  "workflow": "text_to_video",
  "input": {
    "prompt": "A cinematic product shot with deliberate camera movement, natural lighting, and a clear beginning and end."
  },
  "output": {
    "duration_seconds": 1
  }
}'
3

Poll the task

Save id, then poll GET /v1/media/generations/{id} every few seconds until the task is succeeded or failed.

curl https://apineed.com/v1/media/generations/$TASK_ID -H "Authorization: Bearer $API_NEED_API_KEY"

Seedance 2.5 API Endpoint

Create an asynchronous video task, then use its task ID to retrieve status and direct object-storage output URLs.

POST/v1/media/generations
Authorization
Bearer $API_NEED_API_KEY
Content-Type
application/json
HTTP-Referer
optional - your site URL, for rankings
Model
seedance-2.5
GET/v1/media/generations/{id}
Authorization
Bearer $API_NEED_API_KEY
Content-Type
application/json
HTTP-Referer
optional - your site URL, for rankings
Model
seedance-2.5

Parameters

These controls follow APINEED's current DTO and adapter behavior. Unsupported upstream-only fields are intentionally omitted.

NameTypeValuesDescription
modelstringRequiredUse seedance-2.5.
workflowstringtext_to_video, image_to_video, video_referenceSelects a workflow published by the APINEED model catalog.
input.promptstringRequiredDescribe the scene, motion, camera, lighting, and ending state.
input.assetsarrayUp to 2Temporary media URLs with the verified first-frame, last-frame, or reference-video role.
output.duration_secondsinteger1-3600Verified duration seconds control published by the APINEED model catalog.

Seedance 2.5 API questions

Practical answers for choosing, integrating, pricing, and operating the Seedance 2.5 API on APINEED.

What Seedance 2.5 settings has APINEED verified?

The production adapter accepts 4–15 second jobs at 480p or 720p with prompt, reference-image, structured video-reference, aspect-ratio, audio, and watermark controls. [1][3]

Can Seedance 2.5 generate 1080p through APINEED?

No. APINEED currently rejects 1080p for this route before submission. Choose 480p or 720p, or use the separately documented Seedance 2.0 Spicy route for a verified 1080p option. [3]

Which endpoint serves Seedance 2.5 API?

Create a task with POST /v1/video/generations and model seedance-2.5, then poll GET /v1/video/generations/{task_id} until the response reports a terminal status. [3]

How does Seedance 2.5 video-input pricing work?

A request receives the video-input rate only when metadata.content contains a correctly structured video_url item. Prompt-only and image-reference jobs use the no-video-input branch. [1][3]

What discount does APINEED apply to Seedance 2.5?

APINEED charges 97% of the published baseline, which is a 3% discount. The displayed matrix retains separate video-input and no-video-input prices for 480p and 720p. [1][3]

What should production code retain after creating a Seedance 2.5 job?

Persist the APINEED task ID, originating model, submission time, and request context; poll at a controlled interval and preserve upstream error details when a task fails. [3]

How to deploy Seedance 2.5

Connect Seedance 2.5 to your application through the APINEED media API.

1

Add credits

Add APINEED credits before submitting a billable generation.

2

Create an API key

Create and securely store an APINEED API key.

3

Configure the endpoint

Send seedance-2.5 requests to /v1/media/generations.

4

Track the task

Poll the returned task ID and store the completed video URL from outputs[].

More live models