Skip to main content
AI video model
Alibaba
Wan

Wan 3.0 Video

Wan 3.0 combines Standard and Prime tiers with text, first/last-frame, and signed multimodal-reference generation. The reservation is the variant-and-resolution rate multiplied by verified input-video seconds plus reserved output seconds, rounded up once. Intelligent duration reserves 30 output seconds and is unavailable with reference video.

Version
Standard / Prime
Output
480P / 720P / 1080P
Duration
2–30 seconds / intelligent
Credits
16–1,512
Standard and Prime share the same scene and parameter contract
Supports 2–30 seconds plus intelligent duration
Accepts signed images, videos, and audio in reference mode
Supports 480P, 720P, and 1080P with six aspect-ratio choices
30-second overview
Wan
16–1,512 credits
What it does best
A flexible Wan generation entry point for short or long clips, with a higher-priced Prime route when the premium tier is needed.
Best for
Storyboards, product shots, first/last-frame motion, and multimodal direction using short visual and audio references.
Popular searches
How do I use Wan 3.0 Video?How much does Wan 3.0 standard cost?How much does Wan 3.0 Prime cost?

At a glance

What is this model like?

Choose text, frames, or references as one exclusive scene. Reference mode accepts images, videos, and audio, but audio cannot be the only media. Each uploaded asset is checked by true file type, size, geometry, duration, and a model-bound signature before task creation or credit reservation.

What it does best
A flexible Wan generation entry point for short or long clips, with a higher-priced Prime route when the premium tier is needed.
Best for
Storyboards, product shots, first/last-frame motion, and multimodal direction using short visual and audio references.
Why use it on Kyeo AI
Kyeo keeps safety checking server-controlled, verifies every reference before charging, and settles actual usage without hidden extra debits.

Key facts

Quickly assess whether this model fits your use case.

Scenes
Text / first-last frames / references
Reference images
Up to 10
Reference videos
Up to 5; ≤15 seconds total
Reference audio
Up to 5; ≤15 seconds total
Upload formats
JPG / PNG without transparency / WebP / BMP; MP4 / MOV; MP3 / WAV

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.

Wan 3.0
Wan 3.0 Prime
Wan 3.0 AI video

Common questions

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

How do I use Wan 3.0 Video?
How much does Wan 3.0 standard cost?
How much does Wan 3.0 Prime cost?
Which reference media can Wan 3.0 accept?
How does intelligent duration reserve credits?

Selection guide

Use these decision points when choosing a model.

1
Choose standard for cost control

Use standard when the shared scene controls are sufficient and the lower per-second rate matters most.

2
Choose Prime deliberately

Prime uses the same inputs but a higher per-second rate, so compare a short representative shot before scaling a sequence.

3
Use intelligent duration only with enough balance

A duration of -1 reserves the 30-second ceiling. With reference video, choose an explicit duration so input plus output stays within 30 seconds.

Popular comparisons

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

How do Wan 3.0 standard and Prime prices differ?
How do Wan 3.0 and Seedance 2.5 media limits compare?

Model comparison

Compare the current model with alternatives at a glance.

Version and price
Wan 3.0 Video
Standard: 8 / 16 / 32; Prime: 12.2 / 25.2 / 50.4 credits per billed second at 480P / 720P / 1080P.
Seedance 2.5
17–114 credits per input-plus-output or output second, depending on video presence and resolution
Wan 2.7 Video
Fixed 720p / 1080p duration tiers
Multimodal input
Wan 3.0 Video
10 images / 5 videos / 5 audio clips; video and audio each ≤15 seconds total
Seedance 2.5
30 images / 10 videos / 10 audio clips; separate 30-second totals
Seedance 2.0 Mini
9 images / 3 videos / 3 audio clips

Practical usage insights

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

Tier choice changes reservation directly

The reservation is the variant-and-resolution rate multiplied by verified input-video seconds plus reserved output seconds, rounded up once. Intelligent duration reserves 30 output seconds and is unavailable with reference video.

Signed references close external-link ambiguity

Kyeo accepts only references uploaded through its verified path, so file type, geometry, byte size, and duration are bound before execution.

Capabilities

Three exclusive scenes

Generate from text, a required first frame with optional last frame, or multimodal references.

Standard and Prime routing

Switch quality tier without changing the public media contract.

Controlled multimodal references

Combine up to 10 images, 5 videos, and 5 audio clips within the signed local limits.

Use cases

Product motion studies

Use first and optional last frames to define a controlled transformation or camera move.

Reference-led campaign clips

Combine visual identity, motion samples, and short sound references for a consistent direction.

Longer text-led shots

Plan an explicit 2–30-second sequence without requiring reference media.

Prompt tips

State the scene progression

Describe subject, action, camera, lighting, and ending state in chronological order.

Assign one role per reference

Explain which asset defines identity, motion, composition, or sound to reduce conflicts.

Budget intelligent duration as 30 seconds

Treat -1 as a maximum reservation, not as a low-cost default.

Why choose it

One contract covers standard and Prime.
Supports text, frame, and multimodal-reference workflows.
Output duration reaches 30 seconds.
Every reference is verified before credits are reserved.

What to know first

Reference videos and audio each have a 15-second aggregate limit.
A reference video cannot be combined with intelligent output duration; explicit input plus output must stay within 30 seconds.
PNG files with transparency are rejected.
File and arbitrary-link shortcuts are not available in the current verified path.

Adjustable parameters

Quickly assess whether this model fits your use case.

Generation scene
seedance_scene
Optional
Parameter type: Select
Default: text
text
frames
reference
Version
variant
Optional
Parameter type: Select
Default: standard
standard
prime
Resolution
resolution
Optional
Parameter type: Select
Default: 1080P
480P
720P
1080P
Aspect ratio
aspect_ratio
Optional
Parameter type: Select
Default: adaptive
adaptive
16:9
4:3
1:1
3:4
9:16
Duration
duration
Optional
Parameter type: Integer
Default: 5 seconds
-1 for intelligent, or 2–30 seconds
Generate audio
audio
Optional
Parameter type: Boolean
Default: on
true
false
Seed
seed
Optional
Parameter type: Integer
0–2,147,483,647

Credit usage

16–1,512 credits

Standard: 8 / 16 / 32; Prime: 12.2 / 25.2 / 50.4 credits per billed second at 480P / 720P / 1080P. The reservation is the variant-and-resolution rate multiplied by verified input-video seconds plus reserved output seconds, rounded up once. Intelligent duration reserves 30 output seconds and is unavailable with reference video.

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

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.

AI video model
16–150 credits

Wan 2.7 Video

Wan 2.7 supports text-to-video, first-frame or first-and-last-frame image-to-video, and video editing. It currently offers 720p and 1080p, so compare media requirements, duration, framing, and estimated credits together.

AI video model
16 credits/second at 720p; 24 credits/second at 1080p

Sources

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

Source note

This page covers the released Wan 3.0 standard and Prime contracts verified for Kyeo. Kyeo supports text, first/last-frame, and signed multimodal-reference input; file and link shortcuts remain unavailable until they can use the same local verification boundary.

Last updated: 2026-08-28
Wan 3.0 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.

Current Wan 3.0 site specifications
Platform

Used to verify local upload limits, signed-reference checks, preauthorization, and actual settlement.