Skip to main content
AI video model
ByteDance
Seedance

Seedance 2.0 Mini

Seedance 2.0 Mini combines text, first/last-frame, and multimodal-reference generation in one entry point. Reference video enters the pre-task pricing formula, so both media duration and output duration must be verified before task creation.

Scene
Text / first-last frames / references
Output
480p / 720p
Duration
4–15 seconds
Credits
16–150
Three mutually exclusive scenes prevent fields from being mixed
Accepts up to 9 images, 3 videos, and 3 audio clips
Reference-video duration is included in preauthorization
Web search is available only in text mode
30-second overview
Seedance
16–150 credits
What it does best
A video generation option with a lower-resolution tier and controlled multimodal media limits.
Best for
A good fit for multi-asset storyboards, short clips with audio references, and cost-sensitive concept testing.
Popular searches
How do I use Seedance 2.0 Mini?How much does Seedance 2.0 Mini cost?How many videos can Seedance 2.0 Mini accept?

At a glance

What is this model like?

First/last-frame input and reference media are mutually exclusive. Reference videos and audio clips must each be 2–15 seconds, with separate aggregate limits of 15 seconds. Images and videos are also checked for dimensions, aspect ratio, pixel count, and FPS, and noncompliant input is rejected before credits are reserved.

What it does best
A video generation option with a lower-resolution tier and controlled multimodal media limits.
Best for
A good fit for multi-asset storyboards, short clips with audio references, and cost-sensitive concept testing.
Why use it on Kyeo AI
The upload path verifies true formats and media metadata, while trusted durations produce a preauthorization that does not underestimate the request.

Key facts

Quickly assess whether this model fits your use case.

Reference images
Up to 9
Reference videos
Up to 3; ≤15 seconds total
Reference audio
Up to 3; ≤15 seconds total
Web search
Text mode only

Also known as

The same model may appear under different names across documentation and community discussions; this list keeps them easy to search and compare.

Seedance 2 Mini
Seedance 2.0 Mini
Seedance 2.0 Mini multimodal

Common questions

These practical questions focus on the task, input conditions, and result requirements you should confirm before choosing.

How do I use Seedance 2.0 Mini?
How much does Seedance 2.0 Mini cost?
How many videos can Seedance 2.0 Mini accept?
What image formats does Seedance 2.0 Mini support?
How does web search work in Seedance 2.0 Mini?

Selection guide

Use these decision points when choosing a model.

1
Choose Mini for short, multi-asset work

Mini provides a compact multimodal workflow when both reference-video and reference-audio totals stay within 15 seconds.

2
Compare 2.5 when limits are too low

If you need more media or a 30-second boundary, compare Seedance 2.5.

3
Verify media geometry first

Reference videos must satisfy not only format and duration rules, but also pixel-count, dimension, and FPS contracts.

Popular comparisons

Compare common alternatives on the same task to clarify differences in inputs, controls, and cost.

How do the duration and media limits of Seedance 2.0 Mini and Seedance 2.5 differ?
How does Seedance 2.0 Mini pricing compare with MiniMax H3?

Model comparison

Compare the current model with alternatives at a glance.

Reference-media limits
Seedance 2.0 Mini
9 images / 3 videos / 3 audio clips
Seedance 2.5
30 images / 10 videos / 10 audio clips
MiniMax H3
9 images / 3 videos / 3 audio clips

Practical usage insights

Practical guidance based on public sources, current on-site limits, and representative tasks.

Video input changes the rate

A reference video adds billable duration and switches the request to the corresponding video-input per-second rate.

More modalities require stricter upfront validation

Signed metadata binds upload results to the user, model, and media URL, reducing the risk of client-supplied duration being forged.

Capabilities

First/last-frame generation

Use a first frame or both first and last frames to guide the clip's visual start and finish.

Multimodal references

Combine images, video, and audio as reference signals.

Verifiable pricing

Verified reference-video duration is included in preauthorization through a signed token.

Use cases

Multi-asset concept clip

Use a character image, motion video, and mood audio to test a direction quickly.

First-to-last-frame transition

Plan a short transition between two key frames.

Vertical content draft

Choose 9:16 and limit the number of events for a mobile-first clip.

Prompt tips

Assign each asset a role

State whether each media type supplies the character, motion, camera, or sound reference.

Keep one timeline

Describe actions in sequence within 4–15 seconds instead of stacking conflicting events.

Use search only with text

Do not submit first/last frames or reference media when web search is required.

Why choose it

Supports a complete range of multimodal reference types.
Input counts and geometry boundaries are explicit.
Separate video-input and no-video pricing formulas are transparent.
Media authenticity is validated before task creation.

What to know first

Reference-video and reference-audio totals are each capped at 15 seconds.
Maximum output resolution is 720p.
Web search cannot be used with first/last-frame or reference scenes.
Automated checks do not replace human copyright and safety review.

Adjustable parameters

Quickly assess whether this model fits your use case.

Generation scene
seedance_scene
Optional
Parameter type: Select
Default: text
Text
First frame / first and last frames
Multimodal references
Resolution
resolution
Optional
Parameter type: Select
Default: 720p
480p
720p
Video duration
duration
Optional
Parameter type: Integer
Default: 5 seconds
4–15 seconds

Credit usage

16–150 credits

Without a reference video, billing uses output seconds. With reference video, it uses verified total reference-video duration plus output duration. The 480p rate is 3.8 credits per second without video and 2.4 with video; the 720p rate is 8.2 without video and 5 with video. Kyeo rounds the total up and settles against actual consumption after completion.

Budget tip

Before batch generation, run an A/B test with the same assets across the current model and alternatives to validate quality and avoid wasting credits.

FAQ

Related models

Compare these similar candidates before deciding.

Seedance 2.5

Seedance 2.5 combines a 30-second boundary, higher media counts, and video-editing reference input in one strict scene contract. Automatic duration is not underestimated at the 5-second default; Kyeo preauthorizes against the 30-second maximum.

AI video model
102–4,110 credits

MiniMax H3

MiniMax H3 offers 768P and 2K output with text, first/last-frame, and image/video/audio reference scenes. Reference-video duration and every input image from image 6 onward affect pre-task authorization.

AI video model
64–812 credits

Kling 3.0 Turbo

Kling 3.0 Turbo targets 3–15-second video generation. It uses text mode when no image is uploaded and switches to image-to-video when one valid image is supplied. Resolution and duration directly determine the preauthorization amount.

AI video model
54–338 credits

Sources

Content is based on model documentation, feature references, and the settings available on Kyeo AI.

Source note

Explore Seedance 2.0 Mini: 480p or 720p, 4–15 seconds, first/last frames, up to 9 images, 3 videos, and 3 audio clips, 16–150 credits, and upload limits. Without a reference video, billing uses output seconds. With reference video, it uses verified total reference-video duration plus output duration. The 480p rate is 3.8 credits per second without video and 2.4 with video; the 720p rate is 8.2 without video and 5 with video. Kyeo rounds the total up and settles against actual consumption after completion.

Last updated: 2026-08-28
Seedance 2 Mini current interface notes
Platform

Used to check the input modes, visible controls, media limits, and availability of the version currently connected to this site. The page and workbench show the actual options.