ByteDance

Seedance 2.5

VIDEO

Doubao Seedance 2.5 video generation model supporting silent and audio video generation from text prompts, first-frame images, first-and-last-frame images, and multimodal references including images, videos, and audio.

seedance-2.5

PROVIDERSByteDanceENDPOINTSvideo-generationCAPABILITIEStext_to_video · image_to_video · video_to_video

PRICING

Pricing is based on whether a reference video is used:

Scenario	480p	720p
Without reference video	$0.09374 / sec	$0.20401 / sec
With reference video	$0.05643 / sec	$0.12671 / sec

Without reference video: generated video duration × unit price
With reference video: (reference video duration + generated video duration) × unit price

Seedance 2.5 API

Seedance 2.5 API: Generate 30s AI Videos on JustAPI

Generate videos up to 30 seconds, use up to 50 image, video, and audio references, edit or extend existing footage, and create video with audio—all through the Seedance 2.5 API on JustAPI.

Start building Browse models

Seedance 2.5 AI Video Demo

See how Seedance 2.5 turns a creative direction into a complete cinematic scene.

How to Use the Seedance 2.5 API on JustAPI

Connect to Seedance 2.5 through JustAPI in three steps. Create your account, generate an API key, and send your first video generation request.

Create Your Account

Generate Your API Key

Create an API key from your dashboard and use it to authenticate your Seedance 2.5 requests.

Make Your Request

Add a prompt or reference images, videos, and audio, then set options such as duration, resolution, aspect ratio, and audio generation. Send the request to start generating your video.

What’s New in Seedance 2.5?

Build longer, more flexible video workflows with multimodal references, frame controls, video editing, and audio generation.

Generate Videos Up to 30 Seconds

Give your scenes more time to unfold with videos up to 30 seconds long. Create fuller product demos, conversations, ads, and story moments without piecing together a series of short clips.

Up to 50 Multimodal References

Use up to 30 images, 10 videos, and 10 audio files in a single generation. Combine references for characters, products, movement, environments, voices, music, and style to communicate your idea with more than a prompt alone.

Multi-Round Video Extension

Continue building from an existing video instead of starting over. Multi-round extension helps carry the story, action, atmosphere, and main subjects into the next segment while keeping the overall sequence more consistent.

Clay Render References

Start with a simple 3D clay render to map out where characters stand, how they move, and where the camera should be. Seedance 2.5 can use that structure as a reference when creating the finished scene, giving you more direction over complex compositions.

What Can You Build with the Seedance 2.5 API?

From campaign videos to short-form stories, use Seedance 2.5 to turn your ideas and reference materials into content for different creative needs.

Ads and Marketing

Produce more campaign-ready video from the same pool of creative assets. Turn product shots, brand visuals, motion references, and audio into launch videos, promotional ads, and new creative variations for different campaigns.

Seedance 2.5 lipstick campaign video with product messaging and playback controls

Short Dramas and Stories

Build dialogue, character scenes, short dramas, and story sequences with more space for the narrative to unfold. Extend scenes when needed while keeping the story and main subjects connected.

Seedance 2.5 short drama video with dialogue captions and social engagement controls

E-commerce and Product Videos

Move beyond static product photos. Create videos that put products in context, highlight key details, or show how they could look in use—ideal for storefronts, paid ads, and product promotions.

Seedance 2.5 hand cream product video promoting a spring sale

Social Media Content

Create videos for TikTok, Instagram, YouTube Shorts, and other social channels. From trend-inspired clips to branded stories, turn one creative direction into fresh content for different posts and campaigns.

Seedance 2.5 social video example formatted for YouTube, TikTok, Instagram, and Snapchat

Why Use Seedance 2.5 Through JustAPI?

Get easier access to Seedance 2.5 with affordable pricing, fewer restrictions on real-person references, and one API that covers multiple Seedance models.

Affordable Seedance 2.5 API Pricing

Access Seedance 2.5 at a more affordable rate for everyday and high-volume video generation. Clear usage-based pricing makes it easier to plan costs and keep spending under control as you create more videos.

Upload Real-Person References Without a Material ID

Use real-person images or videos directly as references without creating a separate material ID first. This removes an extra preparation step and makes it easier to create videos with characters, presenters, or other real people.

One Seedance API for 2.0, 2.0 Mini, and 2.5

Access Seedance 2.0, Seedance 2.0 Mini, and Seedance 2.5 through the same API. Switch between models based on quality, speed, cost, or creative needs without building and maintaining a separate integration for each version.

Seedance 2.5 vs Seedance 2.0

Both models support multimodal video creation with image, video, and audio references. Seedance 2.5 builds on that foundation with longer single generations, more reference inputs, stronger long-form continuity, and more precise creative control.

Parameter	Seedance 2.5	Seedance 2.0
Max Video Duration	Up to 30 seconds per generation	Up to 15 seconds per generation
Multimodal References	Up to 30 images, 10 videos, and 10 audio files	Up to 9 images, 3 videos, and 3 audio files
Long-Form Creation	Multi-round extension with stronger continuity across longer sequences	Video extension supported
Reference Control	Expanded clay render, motion, and creative referencing	Multimodal reference generation
Video Editing	More precise editing, including timestamp-level control	Multimodal video editing
Best For	Longer stories, complex scenes, reference-heavy projects, and more controlled production	Shorter videos and general multimodal video creation

Seedance 2.5 API FAQ

What is the Seedance 2.5 API?

The Seedance 2.5 API lets you add ByteDance’s multimodal video generation model to your own product or workflow. It supports text prompts, images, videos, and audio references for creating and controlling video content. Seedance 2.5 is available through JustAPI.

How do I access the Seedance 2.5 API through JustAPI?

Create a JustAPI account, generate an API key from your dashboard, and use the key to authenticate your requests. You can then select Seedance 2.5 as the model and start sending generation tasks.

How much does the Seedance 2.5 API cost?

Seedance 2.5 uses usage-based pricing on JustAPI. Your final cost depends on the generation configuration. Check the live pricing on this page for the current rate before sending production requests.

Which parameters are required for a Seedance 2.5 request?

The model and duration parameters are required. A prompt is required for text-to-video, but it can be optional when your uploaded media already provides enough information to define the task.

Does using the same seed produce the same video every time?

Not necessarily. You can set a seed value from 0 to 2147483647 to improve reproducibility, but the same seed does not guarantee an identical result on every generation.

Can I generate videos without a watermark?

Yes. The watermark parameter is available and is set to false by default.

Parameter	Type	Default	Range / values	Description
`model` *	`string`	—	—	No description available.
`prompt`	`string`	—	—	Text prompt describing the video to generate or the edit/extension to apply. Required for text-to-video; optional when reference media fully defines the task.
`content`	`array`	—	Supported items: 0-30 images, 0-10 videos, 0-10 audios. Image item: {type: image_url, image_url: {url: string}, role: first_frame \| last_frame \| reference_image}. Video item: {type: video_url, video_url: {url: string}, role: reference_video}. Audio item: {type: audio_url, audio_url: {url: string}, role: reference_audio}. URLs may be direct URLs or supported asset:// references.	Reference content array. For text-to-video, omit content and provide prompt. For first-frame image-to-video, provide one image with role first_frame. For first-and-last-frame image-to-video, provide one first_frame image and one last_frame image. For multimodal reference generation, provide up to 30 reference images, up to 10 reference videos, and up to 10 reference audios. Audio-only input is supported. For video editing or extension, provide the source video with role.
`duration` *	`integer`	`5`	4 ~ 30 seconds	Target video duration in seconds. Seedance 2.5 supports continuous video output up to 30 seconds.
`resolution` *	`enum`	`720p`	480p · 720p	Output video resolution.
`ratio`	`string`	`16:9`	For example: 16:9, 9:16, 1:1, 4:3, 3:4	Output aspect ratio. For image-to-video tasks, the service may follow the input image ratio when ratio is omitted.
`generate_audio`	`boolean`	`false`	—	Whether to generate audio for the target video. Set true for audio video output, or provide reference_audio items to guide the soundtrack/audio style.
`watermark`	`boolean`	`false`	—	Whether to add a watermark to the generated video.
`seed`	`integer`	—	0 ~ 2147483647	Random seed for reproducibility. The same seed improves reproducibility but does not guarantee identical outputs every time.