Seedance AI Seedance 2.5New Seedance 2.0 Seedance 2.0 Mini Seedance 1.5 Pro Seedance 1.0 Pro Seedance Prompt Guide Seedance Explore Seedance Blog

Seedance AI

Seedance AI Video Generator

Create with Seedance AI using text, images, audio, and video references. Explore the complete model family below and choose the right workflow for your next video.

UploadUpload Media*

Click to upload assets

Max 50 files (img ≤30, vid ≤10, audio ≤10)

Prompt *Describe your vision

Model

Aspect Ratio

Resolution

Duration5s

4s30s

Content Policy Notice

Any non-compliant generation will fail. Due to upstream restrictions, real human faces (selfies, portraits, celebrities) and copyrighted content are not supported. Violent and NSFW content is also rejected. Please use illustrations, anime characters, or AI-generated faces instead.

Compare models

Which Seedance AI model should you use?

Start with the kind of video you need to make. The table highlights each model's role, supported workflow, and clearest advantage.

Compare	2.5Current	2.0	2.0 Mini	1.5 Pro	1.0 Pro
Best fit	Maximum creative control	Cinematic production	Fast short-form creation	Native audio-visual creation	Core cinematic generation
Inputs	Text, image, video, audio	Text + up to 12 media references	Text + up to 12 media references	Text + start/end images	Text + start/end images
Upload resources	50 total · 30 images / 10 videos / 10 audio	12 total · up to 9 images / 3 videos / 3 audio	12 total · up to 9 images / 3 videos / 3 audio	First frame + optional end frame	First frame + optional end frame
Audio	Native audio-video output	Audio-visual generation	Reference audio support	Native speech and sound	Visual generation
Output	4–30s · 480p–1080p	4–15s · 480p–4K Standard	4–15s · 480p–720p	5 or 10s · 480p–1080p	5 or 10s · 480p–1080p
Key advantage	50 references + local re-draw	Multi-shot storytelling	Fast iteration	Performance + lip sync	Simple, proven workflow
Model page	Try 2.5	Try 2.0	Try 2.0 Mini	Try 1.5 Pro	Try 1.0 Pro

Available settings can vary by generation mode. Open a model page to review its current controls before creating.

Model collection

Explore every Seedance AI model

Each generation has a different strength. Review the advantages and ideal use cases below, then open the dedicated model page to see examples and start creating.

2.5

Model 01 · Seedance archive

Current model

Seedance 2.5

Best for

Complex campaigns, longer narrative clips, and reference-heavy productions.

The most capable Seedance workflow for longer, high-fidelity videos built from a large set of creative references.

Try Seedance 2.5

Clip length: 4–30 seconds
Output: 480p–1080p
Inputs: Text · image · video · audio
References: Up to 50 files

Upload resources

Upload up to 30 images, 10 videos, and 10 audio files. Images can be up to 30MB, videos 200MB, and audio 15MB each.

Build a complete 30-second shot

Generate a longer single-segment video while preserving the direction and visual flow of the prompt.

Direct with up to 50 references

Combine images, video, and audio to define characters, environments, movement, timing, and sound.

Refine without starting over

Use local re-draw editing to revise a specific region or detail while keeping the rest of the clip intact.

2.0

Model 02 · Seedance archive

Seedance 2.0

Best for

Commercial concepts, story sequences, branded videos, and previsualization.

A production-focused model that combines multi-shot storytelling, reference-driven direction, and cinematic output.

Try Seedance 2.0

Clip length: 4–15 seconds
Output: 480p–4K Standard
Inputs: Text · image · video · audio
References: Up to 12 files

Upload resources

Upload up to 9 images, 3 videos, and 3 audio files within the 12-file total. Fast mode supports 480p and 720p output.

Tell a story across multiple shots

Build connected scenes with more coherent subjects, actions, and visual language from shot to shot.

Direct motion and composition

Use references and natural-language instructions to guide camera movement, staging, and scene rhythm.

Prepare high-resolution outputs

Choose output settings from draft-friendly resolutions through 4K for more production-ready results.

2.0 Mini

Model 03 · Seedance archive

Seedance 2.0 Mini

Best for

Short-form social clips, quick visual tests, and high-volume creative exploration.

A streamlined model for quickly turning prompts and references into short cinematic video concepts.

Try Seedance 2.0 Mini

Clip length: 4–15 seconds
Output: 480p–720p
Inputs: Text · image · video · audio
References: Up to 12 files

Upload resources

The current Mini creator accepts up to 9 images, 3 videos, and 3 audio files within the 12-file total.

Explore more ideas in less time

A speed-focused workflow makes it easier to test prompts, compositions, and creative directions quickly.

Create compact, usable clips

Generate 4-to-15-second videos suited to social posts, ads, transitions, and concept previews.

Stay flexible with references

Use text, images, video, and audio references while retaining camera and character controls.

1.5 Pro

Model 04 · Seedance archive

Seedance 1.5 Pro

Best for

Dialogue, performance-led videos, entertainment shorts, and sound-rich scenes.

An audio-visual model designed for expressive performances, synchronized sound, and coherent cinematic sequences.

Try Seedance 1.5 Pro

Clip length: 5 or 10 seconds
Output: 480p–1080p
Inputs: Text · start/end images
Audio: Native audio toggle

Upload resources

Start from a prompt or upload a first-frame image, with an optional end-frame image to guide the transition and final composition.

Generate audio and visuals natively

Create speech, spatial sound effects, and moving images together instead of assembling them separately.

Match speech with performance

Improved lip synchronization and expressive character motion make dialogue-driven scenes feel more connected.

Maintain cinematic continuity

Native multi-shot generation supports a more consistent story, look, and emotional arc across scenes.

1.0 Pro

Model 05 · Seedance archive

Seedance 1.0 Pro

Best for

Straightforward text-to-video, image animation, storyboards, and visual prototypes.

The foundational professional Seedance model for prompt-led and image-led cinematic video generation.

Try Seedance 1.0 Pro

Clip length: 5 or 10 seconds
Output: 480p–1080p
Inputs: Text · start/end images
Audio: Visual generation only

Upload resources

Create from text alone or upload a first-frame image, with an optional end-frame image for a more directed image-to-video result.

Start from text or an image

Turn a written scene into video or animate a source image without needing a complex reference setup.

Create believable movement

Physics-aware motion and cinematic camera controls help shots feel intentional and visually grounded.

Preserve the main subject

Subject-consistency controls help retain recognizable characters and objects across generated shots.