Preview

Seedance 2.0 API

Programmatic access to next-generation multimodal AI video generation — transform images, text, audio, and video references into cinematic, production-ready clips. Publicly launching February 24, 2026.

Overview

What is the Seedance 2.0 API?

The Seedance 2.0 API allows developers to integrate advanced AI video generation directly into their applications, products, or internal workflows. With a simple REST-based interface, you can generate high-quality videos from text prompts, images, and multimodal references — enabling use cases like AI video creation tools, marketing automation, creative prototyping, and AI-powered storytelling.

New to Seedance 2.0? Learn more about its core capabilities and what makes it different in our What is Seedance 2.0 overview.

Public Launch

Feb 24, 2026

Features

Planned API Capabilities

At launch, the Seedance 2.0 API is designed to support the following. Feature availability may evolve during the preview phase.

Multimodal Engine

Upload images, videos, audio, and text — each can serve as either the subject to edit or a reference to draw from. Reference anything: motion, effects, style, camera work, characters, scenes, or sound.

Foundational Quality Leap

More realistic physics, smoother and more natural motion, more precise prompt comprehension, and more consistent style across every frame — a comprehensive evolution at the core level.

Consistency & Replication

Faces, outfits, product details, and scene styles stay rock-solid across every frame. Faithfully replicates complex camera work, choreography, and cinematic sequences from any reference.

Creative Continuity

Feed existing videos directly — target specific segments, actions, or pacing for precise edits without regenerating from scratch. Extend clips with strong creative coherence and seamless shot continuity.

Audio-Visual Sync

More accurate timbres and more realistic sound than ever. Supports beat-synced generation — align motion, cuts, and transitions precisely to the rhythm of your music track.

Dynamic Action

Fight sequences, fast chases, acrobatic stunts — high-intensity scenes with physically grounded body dynamics, believable collisions, and responsive camera tracking.

Specifications

Input Parameters

Seedance 2.0 supports four input modalities that can be freely mixed in a single request.

Image

FormatsJPEG, PNG, WebP, BMP, TIFF, GIF
Max countUp to 9 images
File size< 30 MB each

Video

FormatsMP4, MOV
Max countUp to 3 videos, 2–15s total
Resolution480p (640×640) – 720p (834×1112)
File size< 50 MB each

Audio

FormatsMP3, WAV
Max countUp to 3 clips, total ≤ 15s
File size< 15 MB each

Text & Output

Text inputNatural language prompts
Duration4–15 seconds, freely selectable
Audio outputBuilt-in sound effects & music

Combined limits: Mixed input total cap is 12 files per request. Prioritize assets that impact visuals or rhythm the most, and balance file counts across modalities accordingly.

Pricing

API Pricing

Official pricing will be announced at public launch on February 24, 2026. Preview pricing may differ from final public rates.

FAQ

Frequently Asked Questions

Is the Seedance 2.0 API available now?

The API is currently in preview. Public access will be available on February 24, 2026. Join the waitlist to get notified when it launches.

What input types does the API support?

Seedance 2.0 supports four input modalities: images, videos, audio clips, and text prompts. You can mix and match these freely in a single request, with a combined limit of 12 files per generation.

What are the generation limits?

Video generation duration is freely selectable from 4 to 15 seconds. Mixed inputs support up to 12 files total per request. Video references must be 2–15 seconds combined and under 50 MB each. Audio clips total duration must not exceed 15 seconds.

Will pricing change after launch?

Official pricing will be announced at public launch. Preview pricing may differ from the final public pricing. Join the waitlist to receive pricing details first.

Are there any content restrictions?

Currently, uploading materials containing realistic human faces is not supported. This restriction helps ensure responsible AI usage and privacy compliance. The policy may evolve as the platform matures.

Seedance 2.0 Showcase

Videos Generated with Seedance 2.0

Discover stunning AI-generated videos created with Seedance 2.0. From photography to architecture, people to creative projects - see how Seedance 2.0 transforms ideas into cinematic videos.

Start Building with Seedance 2.0

Generate videos directly and explore Seedance 2.0 workflows today.

Create Now