ByteDance/ VideoAvailable

Seedance 2.5 API

seedance-2.5

Native 30-second single-shot video with up to 50 multimodal references, 4K output, and region-level editing.

Text to video
Image to video
Native audio

Try in playground Read API docs

Playground

Configure a request, inspect the exact JSON, then run it with your account.

API key

Loading keys…

Prompt

ResolutionDurationAspect ratioStarting image URLReference image URLreference_videosgenerateAudioenable_web_search

Response

The API response will appear here.

WHY Seedance 2.5

Production-grade image capabilities

What makes Seedance 2.5 a strong choice for high-fidelity, text-aware image workflows.

Native 30-second single-shot video

Generate continuous video up to 30 seconds in one pass — no stitching, no sequence breaks — well beyond the 15–20s ceiling of peer commercial models.

50 multimodal references

Feed up to 50 reference inputs across image, video, text, and audio at once — the highest reference limit among commercial video models — for precise control over characters, scenes, style, and sound.

Native 4K output

Produce crisp 4K video natively, ready for large displays, product films, and high-DPI social without upscaling artifacts.

Region-level editing

Edit specific regions of a clip while preserving overall consistency — swap a product, a prop, or a background element without re-generating the whole shot.

Multimodal audio-video joint generation

A unified architecture ingests text, image, audio, and video together and emits synchronized audio-visual output in a single task.

TECHNICAL SPECS

Supported parameters

Inputs, supported configurations, and request parameters for the Seedance 2.5 API.

Max duration: 30 seconds (single shot)
Resolutions: 480p · 720p · 1080p · 4K
Reference inputs: Up to 50, multimodal
Audio: Synchronized, jointly generated
Editing: Region-level, consistency-preserving
Provider: ByteDance (Doubao Seedance 2.5)
Endpoint: POST /v1/generations

MODEL WORKFLOWS

What to build with Seedance 2.5

Choose a task that matches the live route capabilities, then adapt one of these prompt starters.

Best-fit tasks

Turn a complete shot brief into a generated video sequence.
Animate a supplied image while preserving its subject and visual direction.
Generate an audio-capable result on supported video routes.

Prompt starters

Replace the brackets

Starter

Create one [duration]-second shot of [subject] in [setting]. The subject [action]. Camera: [movement and framing]. Lighting: [direction and mood]. Keep motion physically coherent and the shot continuous.

Starter

Animate the supplied image into a [duration]-second shot. Preserve [subject details]. Add [subject motion] and [camera motion]. Keep the original composition, lighting, and visual identity stable.

MODEL SELECTION

Choose Seedance 2.5 when the job fits

Compare nearby live routes by task, supported capability, and current entry price before you lock a model ID into production.

Model	Best for	Key capabilities	Entry price	Availability	Action
Seedance 2.5 ByteDance · current page	Turn a complete shot brief into a generated video sequence.	Text to video · Image to video · Native audio	$0.12 / sec	Available	Selected
Seedance 1.5 Pro ByteDance	Turn a complete shot brief into a generated video sequence.	Text to video · Image to video	$0.03 / sec	Available	View route
Seedance 2.0 Fast ByteDance	Turn a complete shot brief into a generated video sequence.	Text to video · Image to video	$0.09 / sec	Available	View route
Seedance 2.0 Pro ByteDance	Turn a complete shot brief into a generated video sequence.	Text to video · Image to video · Multi-reference	$0.08 / sec	Available	View route

Capabilities and starting prices are a current PixMind route snapshot. The accepted request contains the final configuration and billing quote.

GET STARTED

From playground to production

Create a scoped key

Create an API key with the permissions required by this model.

Send the request

Use the model ID and endpoint shown above from your backend service.

Track the task

Read language-model output from the response. For image and video requests, use the returned task ID to retrieve the result.

FAQ

Seedance 2.5 API questions

What makes Seedance 2.5 different from Seedance 2.0?+

Seedance 2.5 adds native 30-second single-shot video, up to 50 multimodal references, and region-level editing — so you can generate longer, more controllable clips in one pass instead of stitching multiple 5–15s segments.

How do the 50 multimodal references work?+

Pass up to 50 reference inputs across image, video, text, and audio in one request. They jointly guide characters, scenes, style, and sound, keeping the output consistent with your brand and creative direction.

Can I edit part of a video without regenerating it?+

Yes. Region-level editing lets you change a specific area — a product, a prop, a background element — while the rest of the clip keeps its motion, lighting, and identity.

Is audio generated together with the video?+

Yes. Seedance 2.5 uses a unified audio-video architecture, so synchronized audio is generated in the same task when the selected mode supports it.

What model ID should I use for the Seedance 2.5 API?+

Use seedance-2.5 in the model field. This is the stable PixMind route identifier shown in the playground request JSON.

Which endpoint does the Seedance 2.5 API use?+

Send POST requests to /v1/generations. Keep your API key on your server and pass it as a Bearer token.

Is the playground using a real API request?+

Yes. When you click Run model, PixMind forwards the displayed payload with the API key you entered. A successful request may create a paid task.

Does PixMind save the API key entered here?+

No. The playground keeps the key only in component memory and sends it for the current request; it is not written to browser storage.

Is API pricing the same as PixMind Studio pricing?+

No. API usage resolves through the apiPlatform pricing role and a separate pay-as-you-go ledger. Studio pricing includes the creation interface and product workflow services.

Why can the final price differ from the starting price?+

The entry price reflects the configuration displayed on this page. Resolution, duration, quality, audio, references, and output count can affect the final task price.

RELATED MODELS