Skip to main content
AI video model
Alibaba
Wan

Wan 2.6

No media selects text-to-video, exactly one image selects image-to-video, and one to three videos select video editing; images and videos cannot be mixed. The default is 1080p, 5 seconds, shot structure Off, and content filtering On.

Workflows
Text / 1 image / 1–3 videos
Credit cost
70–315 credits / request
Prompt limit
1–5,000 / 2–5,000
Upload limit
1 JPG/PNG/WebP · 1–3 MP4/MOV · 10 MB
Wan 2.6 — Route selection priority: No media selects text-to-video, exactly one image selects image-to-video, and one to three videos select video editing; images and videos cannot be mixed.
Wan 2.6 — Valid durations: Text and image inputs support 5, 10, or 15 seconds; video input supports only 5 or 10 seconds.
Wan 2.6 — Six Kyeo prices: 720p at 5/10/15 seconds costs 70/140/210 credits, and 1080p costs 105/210/315 credits. Fifteen seconds is unavailable with video input.
Wan 2.6 — Visible controls and media: Accept exactly one JPG/PNG/WebP image at least 256×256 and no larger than 10 MB, or one to three MP4/MOV videos no larger than 10 MB each. Every file must pass type, size, geometry, and signed-metadata validation before the task starts.
30-second overview
Wan
70–315 credits / request
What it does best
Wan 2.6 — Workflows: No media selects text-to-video, exactly one image selects image-to-video, and one to three videos select video editing; images and videos cannot be mixed.
Best for
Wan 2.6 — Media and safety boundary: Accept exactly one JPG/PNG/WebP image at least 256×256 and no larger than 10 MB, or one to three MP4/MOV videos no larger than 10 MB each. Every file must pass type, size, geometry, and signed-metadata validation before the task starts.
Popular searches
How do I use Wan 2.6?What are the six Wan 2.6 prices?How does Wan 2.6 choose the text workflow?

At a glance

What is this model like?

No media selects text-to-video, exactly one image selects image-to-video, and one to three videos select video editing; images and videos cannot be mixed. Text and image inputs support 5, 10, or 15 seconds; video input supports only 5 or 10 seconds.

What it does best
Wan 2.6 — Workflows: No media selects text-to-video, exactly one image selects image-to-video, and one to three videos select video editing; images and videos cannot be mixed.
Best for
Wan 2.6 — Media and safety boundary: Accept exactly one JPG/PNG/WebP image at least 256×256 and no larger than 10 MB, or one to three MP4/MOV videos no larger than 10 MB each. Every file must pass type, size, geometry, and signed-metadata validation before the task starts.
Why use it on Kyeo AI
Wan 2.6 — Video duration: Text and image inputs support 5, 10, or 15 seconds; video input supports only 5 or 10 seconds. Shot structure defaults to Off, and content filtering defaults to On. Both controls are visible and sent explicitly.

Key facts

Quickly assess whether this model fits your use case.

Category
AI video model
Vendor
Alibaba
Model family
Wan
Runtime
Asynchronous task
Prompt limit
Text: 1–5,000 characters; image/video: 2–5,000 characters
Upload limit
1 JPG/PNG/WebP image (≥256×256, ≤10 MB) or 1–3 MP4/MOV videos (≤10 MB/file)

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.

Alibaba Wan 2.6
Wan 2.6

Common questions

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

How do I use Wan 2.6?
What are the six Wan 2.6 prices?
How does Wan 2.6 choose the text workflow?
Which image specifications does Wan 2.6 require?
How many videos does Kyeo accept for Wan 2.6?
Which Wan 2.6 routes support 15 seconds?
What do Wan 2.6 720p and 1080p cost?
What happens if Wan 2.6 receives an image and video?
What is the minimum Wan 2.6 prompt length?
Wan 2.6 Shot structure
Wan 2.6 Content filter

Selection guide

Use these decision points when choosing a model.

1
Choose one input path first

Choose one input path first: No media selects text-to-video, exactly one image selects image-to-video, and one to three videos select video editing; images and videos cannot be mixed.

2
Filter duration by workflow

Filter duration by workflow: Text and image inputs support 5, 10, or 15 seconds; video input supports only 5 or 10 seconds.

3
Build three separate baselines

Build three separate baselines: The default is 1080p, 5 seconds, shot structure Off, and content filtering On. Compare models with the same prompt, media, duration, and resolution; test returned results instead of assuming marketing claims.

Popular comparisons

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

How does Wan 2.6 differ from Kling 2.6?
Wan 2.6 vs Kling 2.6 input routes and duration
How does Wan 2.6 compare with Hailuo 2.3?
Wan 2.6 vs Hailuo 2.3 media and billing
What separates Wan 2.6 from Seedance 1.5 Pro?
Wan 2.6 vs Seedance 1.5 Pro controls

Model comparison

Compare the current model with alternatives at a glance.

Kyeo input paths
Wan 2.6
Kyeo input paths / Wan 2.6: Text / 1 image / 1–3 videos
Wan 2.7 Video
Separate text, first-last-frame or extension, and video-edit routes
Kling 2.6
Separate text and image routes
Route selection priority
Wan 2.6
Route selection priority / Wan 2.6: No media selects text-to-video, exactly one image selects image-to-video, and one to three videos select video editing; images and videos cannot be mixed.
Wan 2.7 Video
Media selects a 2.7 text, image, or video-edit workflow
Kling 2.6
Image presence switches between text and image routes
Valid durations
Wan 2.6
Valid durations / Wan 2.6: Text and image inputs support 5, 10, or 15 seconds; video input supports only 5 or 10 seconds.
Wan 2.7 Video
Text/image support 5/10/15s; video editing has a separate contract
Kling 2.6
Text and image both use 5/10s
Six Kyeo prices
Wan 2.6
Six Kyeo prices / Wan 2.6: 720p at 5/10/15 seconds costs 70/140/210 credits, and 1080p costs 105/210/315 credits. Fifteen seconds is unavailable with video input.
Wan 2.7 Video
Priced separately by 2.7 workflow, resolution, and duration
Seedance 1.5 Pro
Priced by resolution, duration, and input mode
Visible controls and media
Wan 2.6
Visible controls and media / Wan 2.6: Resolution, Video duration, Shot structure, Content filter
Wan 2.7 Video
Controls and media slots vary by 2.7 workflow
Kling 2.6
Shows Kling-specific duration, ratio, and sound controls

Practical usage insights

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

Wan 2.6 — Media-driven three-workflow selection

Wan 2.6 / Media-driven three-workflow selection: No media selects text-to-video, exactly one image selects image-to-video, and one to three videos select video editing; images and videos cannot be mixed.

Wan 2.6 — Six-entry credit table

Wan 2.6 / Six-entry credit table: 720p at 5/10/15 seconds costs 70/140/210 credits, and 1080p costs 105/210/315 credits. Fifteen seconds is unavailable with video input. Compare models with the same prompt, media, duration, and resolution; test returned results instead of assuming marketing claims.

Capabilities

Media-driven three-workflow selection

Media-driven three-workflow selection: No media selects text-to-video, exactly one image selects image-to-video, and one to three videos select video editing; images and videos cannot be mixed.

Six-entry credit table

Six-entry credit table: 720p at 5/10/15 seconds costs 70/140/210 credits, and 1080p costs 105/210/315 credits. Fifteen seconds is unavailable with video input.

Media and safety boundary

Media and safety boundary: Accept exactly one JPG/PNG/WebP image at least 256×256 and no larger than 10 MB, or one to three MP4/MOV videos no larger than 10 MB each. Every file must pass type, size, geometry, and signed-metadata validation before the task starts. Shot structure defaults to Off, and content filtering defaults to On. Both controls are visible and sent explicitly.

Use cases

Text-only baseline

Text-only baseline: No media selects text-to-video, exactly one image selects image-to-video, and one to three videos select video editing; images and videos cannot be mixed. The default is 1080p, 5 seconds, shot structure Off, and content filtering On.

One-image motion test

One-image motion test: Accept exactly one JPG/PNG/WebP image at least 256×256 and no larger than 10 MB, or one to three MP4/MOV videos no larger than 10 MB each. Every file must pass type, size, geometry, and signed-metadata validation before the task starts. The default is 1080p, 5 seconds, shot structure Off, and content filtering On.

1–3 MP4/MOV — Video duration

1–3 MP4/MOV — Video duration: Accept exactly one JPG/PNG/WebP image at least 256×256 and no larger than 10 MB, or one to three MP4/MOV videos no larger than 10 MB each. Every file must pass type, size, geometry, and signed-metadata validation before the task starts. Text and image inputs support 5, 10, or 15 seconds; video input supports only 5 or 10 seconds.

Prompt tips

Confirm one media path

Confirm one media path: No media selects text-to-video, exactly one image selects image-to-video, and one to three videos select video editing; images and videos cannot be mixed.

Change one output control at a time

Change one output control at a time: 720p at 5/10/15 seconds costs 70/140/210 credits, and 1080p costs 105/210/315 credits. Fifteen seconds is unavailable with video input.

Shot structure / Content filter

Shot structure / Content filter: Shot structure defaults to Off, and content filtering defaults to On. Both controls are visible and sent explicitly. The model does not generate an audio track. Review every result for safety, rights, identity, and publishing quality.

Why choose it

Workflows: No media selects text-to-video, exactly one image selects image-to-video, and one to three videos select video editing; images and videos cannot be mixed.
Credit cost: 720p at 5/10/15 seconds costs 70/140/210 credits, and 1080p costs 105/210/315 credits. Fifteen seconds is unavailable with video input.
Upload limit: Accept exactly one JPG/PNG/WebP image at least 256×256 and no larger than 10 MB, or one to three MP4/MOV videos no larger than 10 MB each. Every file must pass type, size, geometry, and signed-metadata validation before the task starts.
Video duration: Text and image inputs support 5, 10, or 15 seconds; video input supports only 5 or 10 seconds.

What to know first

Wan 2.6 — Filter duration by workflow: Text and image inputs support 5, 10, or 15 seconds; video input supports only 5 or 10 seconds.
Wan 2.6 — Upload limit: Accept exactly one JPG/PNG/WebP image at least 256×256 and no larger than 10 MB, or one to three MP4/MOV videos no larger than 10 MB each. Every file must pass type, size, geometry, and signed-metadata validation before the task starts.
Wan 2.6 — Choose one input path first: No media selects text-to-video, exactly one image selects image-to-video, and one to three videos select video editing; images and videos cannot be mixed.
Wan 2.6 — Shot structure / Content filter: Shot structure defaults to Off, and content filtering defaults to On. Both controls are visible and sent explicitly. The model does not generate an audio track. Review every result for safety, rights, identity, and publishing quality.

Adjustable parameters

Quickly assess whether this model fits your use case.

Resolution
resolution
Optional
Parameter type: select
Default: 1080p
720p
1080p
Video duration
duration
Optional
Parameter type: select
Default: 5
5 seconds
10 seconds
15 seconds (text/image only)
Shot structure
multi-shots
Optional
Parameter type: select
Default: false
Single-shot (false)
Multi-shot (true)
Content filter
content-check
Optional
Parameter type: Dropdown
Default: On
Off
On

Credit usage

70–315 credits / request

720p at 5/10/15 seconds costs 70/140/210 credits, and 1080p costs 105/210/315 credits. Fifteen seconds is unavailable with video input.

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.

Kling 2.6

Kling 2.6 is a two-workflow video service on Kyeo: text requests use text-to-video, while requests with an image use single-image animation. Both routes offer 5/10 seconds and a `sound` toggle; text generation also provides three aspect ratios. The four Kyeo credit tiers are 55, 110, 110, and 220.

AI video model
55-220 credits / request

Hailuo 2.3

Kyeo has six valid combinations: Standard 768P at 6/10 seconds and 1080P at 6 seconds cost 25/40/40 credits; the matching Pro prices are 40/80/70. 1080P does not support 10 seconds. Exactly 1 JPG/PNG/WebP file, up to 10MB.

AI video model
25-80 credits / request

Seedance 1.5 Pro

Seedance 1.5 Pro / Capabilities — Seedance 1.5 Pro on Kyeo AI accepts text alone or up to two images and exposes 480p/720p/1080p, 4–12 seconds, generated audio, six ratios, and a fixed-lens switch. The 7–180-credit cost follows the selected combination; the Pro name and higher credit cost do not prove image quality, stability, or publication readiness.

AI video model
7–180 credits per request

Sources

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

Source note

Reviewed against Alibaba Cloud's official Wan 2.6 series announcement and official video-model overview. The public facts below distinguish text, image, and video inputs, callable durations, local credits, verified media limits, and visible controls; marketing claims are not treated as tested results.

Last updated: 2026-08-27
Alibaba Cloud: Wan 2.6 series launch
Official
View source
Alibaba Cloud: video generation model overview
Official
View source