Q: What is Seedance 2.0 Pro?

Seedance 2.0 Pro is a multi-modal AI video generator powered by the Seedance 2.0 model. It supports text, image, video, and audio inputs for highly controllable video generation with native audio output .

Q: What are Text-to-Video and Media-to-Video?

Text-to-Video (T2V) Describe your scene in the prompt and the AI generates a video from scratch — no uploads needed. Media-to-Video (M2V) Upload reference assets and the AI uses them to guide generation. Use @1, @2, @3 in your prompt to reference uploads. Supported combinations: Text + Images Text + Videos Text + Images + Audio Text + Videos + Audio Text + Images + Videos + Audio ⚠️ Audio cannot be used alone with text — at least one image or video is required. Asset limits: Up to 9 images, 3 videos, 3 audio files (12 total, 50 MB each) Videos and audio: max 15 seconds each When reference images or videos contain real human faces, switch to Real mode; or use the built-in Avatar Library as an alternative

Q: What are the Avatar Library and Material Library?

Avatar Library gives you access to a curated set of virtual human characters — diverse nationalities, ages, and personalities — which you can insert directly as reference assets into your Media-to-Video prompt without any upload. Material Library provides pre-built reference assets organized by category: Clothing — outfits and wardrobe styles Scene — background environments and settings Style — visual styles and art directions Action — motion and gesture references Camera — pre-defined camera movement presets Voice — voice tone and speech style references Both libraries are available in the prompt toolbar. Click the Avatar Library or Material Library button above the prompt input to browse and select — the asset reference is inserted into your prompt automatically.

Q: What's the difference between Fast and Pro models?

Fast (Seedance 2.0 Fast) — Generates quicker at lower cost. 720p–1080p, 4–15 seconds. Pro (Seedance 2.0) — Higher-quality output with finer detail and better motion consistency, at a slightly higher credit cost. Both models support the same durations, resolutions, and aspect ratios.

Q: What durations, resolutions, and aspect ratios are supported?

Duration : 4s, 5s, 8s, 10s, 12s, 15s Resolution : 720p, 1080p Aspect Ratio : 16:9, 9:16, 1:1, 4:3, 3:4, 21:9 All options are available for both Fast and Pro models.

Q: Can I use outputs commercially?

Yes — with a Commercial License , all generated content copyrights belong to you. Plans that include Commercial License: All Annual plans (Starter, Pro, Studio) Monthly Studio plan How to get your license certificate: Subscribe to an eligible plan Go to My License page Download your PDF certificate — it includes your license number, validity period, and granted rights With the certificate, you can freely use outputs for client work, marketing, advertising, social media, and resale.

Q: What should I do if a task fails?

Task failures are usually caused by: Real person content — using photos or videos of real people may trigger content moderation in Standard mode. Switch to Real mode, or use the Avatar Library as an alternative Overly strict content boundaries — when Standard mode blocks too aggressively, try Wild mode (looser content moderation) Policy violations — prompts containing copyrighted characters, explicit, or violent content will be rejected Peak congestion — the API may queue requests during peak hours; try again later What to try: References include real people → switch to Real mode Content blocked too strictly → switch to Wild mode Revise your prompt to avoid restricted content Try again at a different time Credits are automatically refunded for failed tasks.

Q: What is Seed?

Seed is a number that controls the randomness of video generation. By default, every generation uses a random seed, so each result is unique. If you set a fixed seed (e.g. 42) and use the same prompt, settings, and input, you'll get the same result every time. This is useful for: Reproducing a result you liked Making small prompt adjustments while keeping the visual style consistent A/B testing different prompts with the same seed You can find the Seed input in Advanced Settings . Leave it empty for random generation. After a video is generated, the actual seed used is displayed — you can copy it for future use. Seed is available in Standard and Wild modes only.

Q: What are Continue from Last Frame and Continue from Video?

These features let you create continuous stories by using the output of one generation as the starting point for the next. Continue from Last Frame Uses the last frame of your generated video as the first frame of a new generation (Image-to-Video mode). This creates seamless visual continuity between clips — perfect for building multi-scene narratives. Continue from Video Uses the entire generated video as a reference asset for a new generation (Media-to-Video mode). This preserves style, motion, and character consistency while giving you more creative freedom. Both buttons appear automatically after a video is generated. Click one, write a new prompt for the next scene, and generate.

Question 1

What is Seedance 2.0 Pro?

Accepted Answer

Seedance 2.0 Pro is a multi-modal AI video generator powered by the Seedance 2.0 model. It supports text, image, video, and audio inputs for highly controllable video generation with native audio output.

Question 2

What are Text-to-Video and Media-to-Video?

Accepted Answer

Text-to-Video (T2V)
Describe your scene in the prompt and the AI generates a video from scratch — no uploads needed.

Media-to-Video (M2V)
Upload reference assets and the AI uses them to guide generation. Use @1, @2, @3 in your prompt to reference uploads.

Supported combinations:

Text + Images
Text + Videos
Text + Images + Audio
Text + Videos + Audio
Text + Images + Videos + Audio

⚠️ Audio cannot be used alone with text — at least one image or video is required.

Asset limits:

Up to 9 images, 3 videos, 3 audio files (12 total, 50 MB each)
Videos and audio: max 15 seconds each
When reference images or videos contain real human faces, switch to Real mode; or use the built-in Avatar Library as an alternative

Question 3

What are the Avatar Library and Material Library?

Accepted Answer

Avatar Library gives you access to a curated set of virtual human characters — diverse nationalities, ages, and personalities — which you can insert directly as reference assets into your Media-to-Video prompt without any upload.

Material Library provides pre-built reference assets organized by category:

Clothing — outfits and wardrobe styles
Scene — background environments and settings
Style — visual styles and art directions
Action — motion and gesture references
Camera — pre-defined camera movement presets
Voice — voice tone and speech style references

Both libraries are available in the prompt toolbar. Click the Avatar Library or Material Library button above the prompt input to browse and select — the asset reference is inserted into your prompt automatically.

Question 4

What's the difference between Standard, Real, and Wild modes?

Accepted Answer

These three modes map to different generation channels with identical video generation quality; they differ only in content boundaries and throughput characteristics:

Standard — The default option, covers most creation scenarios
Real — Dedicated to reference images or videos containing real human faces
Wild — Looser content moderation (duration, resolution, and other parameter limits are unchanged)

Capability comparison:

Mode	Real faces	Relaxed filter	Stability
Standard	✗	✗	●●●
Real	✓	✓	●○○
Wild	✗	✓	●●○

For normal use, just choose Standard. If your reference material includes real human faces, switch to Real mode. Real mode uses a generation channel optimized for portrait processing, involving additional compliance review and face consistency guarantees, so stability is lower and credit consumption is slightly higher than Standard mode.

Question 5

What's the difference between Fast and Pro models?

Accepted Answer

Fast (Seedance 2.0 Fast) — Generates quicker at lower cost. 720p–1080p, 4–15 seconds.

Pro (Seedance 2.0) — Higher-quality output with finer detail and better motion consistency, at a slightly higher credit cost.

Both models support the same durations, resolutions, and aspect ratios.

Question 6

How are credits calculated and deducted?

Accepted Answer

Credits are calculated per second based on model, resolution, and whether you include video input.

Without video input — credits = output duration × per-second rate
With video input — credits = (input video duration + output duration) × per-second rate (lower rate)

Per-second rates (1x, annual / Max Pack users):

Model	No Video Input	With Video Input
Fast 720p	9 credits/sec	5 credits/sec
Fast 1080p	19 credits/sec	12 credits/sec
Pro 720p	11 credits/sec	7 credits/sec
Pro 1080p	24 credits/sec	15 credits/sec

Examples:

Pro 720p, 5s output, no video = 5 × 11 = 55 credits
Pro 720p, 5s output + 5s video input = (5+5) × 7 = 70 credits

⚠️ Monthly and other non-annual users are charged 1.5× these rates.

When are credits deducted?
Credits are deducted upfront when you submit a task. The exact cost is shown in the submit button before you click. If the task fails, credits are automatically refunded — no action needed on your part.

Question 7

What durations, resolutions, and aspect ratios are supported?

Accepted Answer

Duration: 4s, 5s, 8s, 10s, 12s, 15s
Resolution: 720p, 1080p
Aspect Ratio: 16:9, 9:16, 1:1, 4:3, 3:4, 21:9

All options are available for both Fast and Pro models.

Question 8

Can I use outputs commercially?

Accepted Answer

Yes — with a Commercial License, all generated content copyrights belong to you.

Plans that include Commercial License:

All Annual plans (Starter, Pro, Studio)
Monthly Studio plan

How to get your license certificate:

Subscribe to an eligible plan
Go to My License page
Download your PDF certificate — it includes your license number, validity period, and granted rights

With the certificate, you can freely use outputs for client work, marketing, advertising, social media, and resale.

Question 9

What should I do if a task fails?

Accepted Answer

Task failures are usually caused by:

Real person content — using photos or videos of real people may trigger content moderation in Standard mode. Switch to Real mode, or use the Avatar Library as an alternative
Overly strict content boundaries — when Standard mode blocks too aggressively, try Wild mode (looser content moderation)
Policy violations — prompts containing copyrighted characters, explicit, or violent content will be rejected
Peak congestion — the API may queue requests during peak hours; try again later

What to try:

References include real people → switch to Real mode
Content blocked too strictly → switch to Wild mode
Revise your prompt to avoid restricted content
Try again at a different time

Credits are automatically refunded for failed tasks.

Question 10

Any tips for writing effective Seedance 2 prompts?

Accepted Answer

For best results, be specific and descriptive. A few key guidelines:

Always describe the subject, action, scene, and mood
In M2V mode, use @1, @2, @3 to reference uploaded assets in the prompt
Use the Avatar Library and Material Library to insert pre-built references without uploading
Specify camera movements (e.g. slow push in, overhead tracking shot) for cinematic control
For audio-guided generation, describe voice tone, music genre, or sound design

We have a full interactive guide with real examples: Seedance 2.0 Prompt Guide →

Question 11

What is Seed?

Accepted Answer

Seed is a number that controls the randomness of video generation. By default, every generation uses a random seed, so each result is unique.

If you set a fixed seed (e.g. 42) and use the same prompt, settings, and input, you'll get the same result every time. This is useful for:

Reproducing a result you liked
Making small prompt adjustments while keeping the visual style consistent
A/B testing different prompts with the same seed

You can find the Seed input in Advanced Settings. Leave it empty for random generation. After a video is generated, the actual seed used is displayed — you can copy it for future use.

Seed is available in Standard and Wild modes only.

Question 12

What are Continue from Last Frame and Continue from Video?

Accepted Answer

These features let you create continuous stories by using the output of one generation as the starting point for the next.

Continue from Last Frame
Uses the last frame of your generated video as the first frame of a new generation (Image-to-Video mode). This creates seamless visual continuity between clips — perfect for building multi-scene narratives.

Continue from Video
Uses the entire generated video as a reference asset for a new generation (Media-to-Video mode). This preserves style, motion, and character consistency while giving you more creative freedom.

Both buttons appear automatically after a video is generated. Click one, write a new prompt for the next scene, and generate.

Capability	Seedance 2.0	Seedance 2.5	Seedance 2.0 Mini
Max clip length	4-15 seconds	Up to 30 seconds	4-15 seconds
Resolution	Up to 4K	Native 4K	Up to 1080p
Reference inputs	Up to 12 files (9 images + video + audio)	Up to 50 files (30 images + 10 videos + 10 audio)	Up to 12 files
Native audio	Yes	Yes	Yes
Best for	Ads, social clips, story scenes	Short films, UHD brand videos, licensed IP work	High-volume drafts at the lowest cost

Capability	Seedance 2.0	Seedance 2.5	Seedance 2.0 Mini
Max clip length	4-15 seconds	Up to 30 seconds	4-15 seconds
Resolution	Up to 4K	Native 4K	Up to 1080p
Reference inputs	Up to 12 files (9 images + video + audio)	Up to 50 files (30 images + 10 videos + 10 audio)	Up to 12 files
Native audio	Yes	Yes	Yes
Best for	Ads, social clips, story scenes	Short films, UHD brand videos, licensed IP work	High-volume drafts at the lowest cost

Seedance 2.0 AI Video Generator

What is Seedance 2.0?

Text, image, video & audio references

Native audio generation

Character & style consistency

Flexible 4-15 second clips

Standard and Pro quality tiers

What can you make with Seedance 2.0?

Short ads and product videos

Social content at scale

Story scenes and previsualization

How to use Seedance 2.0

Describe the shot

Add references

Pick length and resolution

Generate and iterate

Seedance 2.0 vs Seedance 2.5 vs Seedance 2.0 Mini

Seedance 2.0 Pro — FAQ

Explore more