What is Seedance 2.0?

Seedance 2.0 is an advanced AI video generation model available through supported providers. It supports true multi-modal input — combining images, videos, audio, and text in a single prompt to create cinematic videos with reference-driven control, persistent character identity, and native audio-video synchronization.

How is Seedance 2.0 different from competitors like Runway or Pika?

Seedance 2.0 is the only model that supports multi-modal reference input (@images for composition, @videos for camera/motion replication), maintains persistent character identity across scenes, and generates native audio-video sync in a single pass. It also supports up to 15-second video generation in 6 aspect ratios.

What input modes are supported?

Seedance 2.0 accepts four types of input simultaneously: up to 9 images, 3 videos (total ≤15s), 3 audio files (MP3, total ≤15s), and text prompts. The total file limit is 12 files per generation. You can mix and match these inputs for precise creative control.

What video specifications are supported?

Videos can be 4–15 seconds long, generated in 6 aspect ratios (16:9, 9:16, 4:3, 3:4, 21:9, 1:1). The model generates native audio (sound effects, music, and dialogue) synchronized with the video content.

Is there a free trial?

Eligible new users may receive trial credits after account checks. You can also buy credit packs or choose a monthly plan when you need more generations.

When will the full platform launch?

Seedance AI is live now! Sign up to start creating videos immediately. We offer Free, Pro ($29/mo), and Business ($99/mo) plans to fit your creative needs.

Can Seedance 2.0 maintain character consistency across scenes?

Yes. Seedance 2.0 uses persistent character identity technology to lock in facial features, clothing, and body types across different angles, scenes, and lighting conditions. Simply @reference a character image and the model maintains that identity throughout extended sequences — essential for storytelling and commercial production.

What camera movements can Seedance 2.0 replicate?

Seedance 2.0 can precisely replicate real-world camera movements including dolly shots, crane movements, tracking shots, whip pans, and complex multi-axis transitions. Upload a reference video with @video and the model extracts and reproduces the exact camera trajectory in your generated content.

How does Seedance 2.0 compare to Kling, Runway, Sora, and Pika?

Seedance 2.0 is the first model to combine true multi-modal input (images + videos + audio + text), persistent character identity, camera/motion replication from reference videos, and native audio-video sync in a single generation pass. Competitors typically support only text-to-video or image-to-video with limited reference control.

Does Seedance 2.0 support audio and music generation?

Yes. Seedance 2.0 generates native audio synchronized with video content, including sound effects, background music, and dialogue. You can also upload reference audio files (MP3, up to 3 files totaling ≤15s) to guide the audio style, beat synchronization, and voice tone of the generated video.

What is multi-modal input and why does it matter?

Multi-modal input means you can combine multiple types of media — images, videos, audio clips, and text — in a single creative prompt. This matters because it gives you director-level control: use images for character and composition, videos for camera work and motion, audio for rhythm and tone, and text for scene description. No other AI video model offers this level of combined control.

Can I use Seedance 2.0 for commercial projects?

Seedance 2.0 is designed for both personal and commercial use. Content creators, marketing teams, filmmakers, and businesses can use it to produce product ads, social media videos, trailers, and branded content. Specific licensing terms will be available at launch — join the waitlist for early details.

What video editing capabilities does Seedance 2.0 offer?

Beyond generation, Seedance 2.0 supports video extension (seamlessly adding seconds to existing clips), character swapping, element addition/removal, and creative template replication. Think of it as continuous shooting rather than one-off generation — you can iteratively refine and extend your video projects.

Create AI Videos with a Real Face using Seedance 2.0

Put a real human face into any AI video scene with Seedance 2.0 Fast and Standard. Generate videos and Nano Banana images directly from the homepage.

🎬 Available now: Seedance 2.0 video, Nano Banana Pro, and Nano Banana 2 image generation

View Pricing

See What Seedance 2.0 Can Create

Hover to preview motion instantly. The homepage keeps a curated selection, and the full inspiration library lives in the showcase gallery.

Seedance 2.0 at a Glance

The most capable AI video generation model, now available for early access.

4 Input Modes

Image + Video + Audio + Text

Up to 15s

Freely choose 4–15 seconds

6 Aspect Ratios

16:9, 9:16, 4:3, 3:4, 21:9, 1:1

Native Audio

Built-in SFX, music & dialogue

Why Seedance 2.0 Changes Everything

From text to cinematic masterpiece — the most capable AI video model ever created.

Multi-Modal Input

Combine up to 9 images, 3 videos, 3 audio tracks, and text in a single prompt — true director mode.

Reference-Driven Control

@mention reference images for composition and character, @reference videos for camera and motion replication.

Persistent Character Identity

Lock in facial features, clothing, and body types across different angles, scenes, and lighting conditions.

Camera & Motion Replication

Precisely replicate real-world camera movements, dance choreography, action sequences, and complex transitions.

Creative Template Replication

Reproduce complex VFX, costume changes, AE-style intros, and creative templates with stunning fidelity.

Native Audio-Video Sync

Built-in SFX, music, and dialogue generation with beat-synced music and accurate voice tones.

Video Extension & Editing

Seamlessly extend clips, swap characters, add or remove elements — not just generation, but continuous shooting.

Cinema-Grade Output

4–15 second videos with smooth one-take continuity, expressive emotional acting, and 6 aspect ratios.

How It Works

From idea to cinematic video in three steps:

Describe Your Vision

Write a text prompt describing your scene. Optionally add reference images, videos, or audio for precise control.

AI Generates Your Video

Seedance 2.0's Dual-Branch DiT processes your multi-modal input and generates cinema-grade video with synchronized audio.

Download & Share

Preview your video, make adjustments if needed, and download in your preferred format and resolution.

Start Creating with Seedance 2.0

Start creating after signup. Eligible new users may receive trial credits.

Start Creating

Frequently Asked Questions

Everything you need to know about Seedance 2.0