Alibaba · Text to video, second by second

Wan 3.0

Four Things Wan 3.0 Changes

Alibaba's newest video generation model, available on SharkFoto as a pure text-to-video tool. Pick any length from 1 to 15 seconds, any of six framings, and a 480P, 720P or 1080P render - all from a single written prompt.

1080p
Top resolution
1-15s
Clip length
6
Aspect ratios
2,500
Prompt characters
What's new

Four Things Wan 3.0 Changes

01

Length to the Second

Set any whole number of seconds from 1 to 15. Earlier Wan models on SharkFoto only offered fixed 5, 10 and 15 second steps, so a 7-second cut meant generating 10 and trimming. Here you ask for 7 and pay for 7.

02

Six Framings, Including Adaptive

16:9, 9:16, 1:1, 4:3 and 3:4 cover landscape, vertical, square and classic-camera shapes. Adaptive hands the framing decision to the model instead of pinning it to a fixed ratio, which suits prompts that describe a shot rather than a delivery format.

03

A 480P Tier for Drafts

Wan 3.0 adds a 480P option below the usual HD tiers. Because credits scale with resolution, a 480P pass is the cheapest way to test whether a prompt reads the way you intended before committing to a 1080P render.

04

Room for a Real Brief

The prompt field accepts up to 2,500 characters - 500 more than Wan 2.6. That is enough to spell out subject, wardrobe, lighting, lens, camera move and pacing in one go instead of compressing the whole idea into a sentence.

Overview

Model at a glance

Wan 3.0 is the current generation of Alibaba's Wan video model family, released publicly in August 2026. On SharkFoto it runs as a text-to-video model: you write a prompt, choose the shape, resolution and length of the clip, and the model renders the whole thing in one pass. No reference image, no storyboard, no editing timeline.

The practical difference from earlier Wan releases is control granularity. Duration is a continuous 1-15 second range rather than a set of preset buttons, the aspect ratio list grows from two options to six, and a 480P tier sits underneath 720P and 1080P so that iteration costs a fraction of delivery. The prompt window is larger as well, which matters more than it sounds: a video prompt has to carry subject, motion and camera direction all at once.

Wan 3.0 is built for single-prompt generation, so it plans the shot itself and holds subject appearance, spatial layout and motion consistent across the clip. That makes it a good fit for short social cuts, product and marketing b-roll, and quick visual concepts - anywhere the idea exists as a description rather than as footage you already have.

Developer
Alibaba
Model
Wan 3.0
Mode
Text to video
Clip Length
1-15 seconds (any whole second)
Resolution
480P, 720P, 1080P
Aspect Ratios
Adaptive, 16:9, 9:16, 1:1, 4:3, 3:4
Prompt Limit
2,500 characters
Reference Inputs
None - prompt only
Capabilities

Key features

Per-Second Duration Control

Duration is a slider across the full 1-15 second range, and every whole second is a valid request. A 2-second loop for a feed, an 8-second product beat, a 15-second full cut - each is asked for directly rather than trimmed out of a longer render.

Adaptive and Fixed Framing

Five fixed ratios cover the formats delivery actually requires: 16:9 for landscape and YouTube, 9:16 for Reels, Shorts and TikTok, 1:1 for feed posts, 4:3 and 3:4 for the more photographic shapes. Adaptive is the sixth option, for when the prompt should decide the frame.

Three Resolution Tiers

480P, 720P and 1080P are separate renders, not an upscale chain. Credits scale with the tier, so the same prompt can be explored cheaply at 480P and then re-rendered once at 1080P when the wording is settled.

2,500-Character Prompts

A long prompt window lets you direct rather than describe: subject and wardrobe, the environment and time of day, the lighting, the lens, how the camera moves, and what changes between the first and last second of the clip.

Consistency Across the Clip

Wan 3.0 is designed to hold character detail, spatial relationships and motion coherent for the whole duration rather than drifting as the clip runs on, so faces, props and background geometry stay recognisable from the first frame to the last.

Shot Planning from One Prompt

The model handles multi-shot structure natively, so a prompt that describes a small sequence does not have to be split into separate generations and cut together by hand afterwards.

Specs

Technical specifications

Video Output

Resolution480P, 720P or 1080P (default 720P)
Clip Length1-15 seconds, any whole second (default 5s)
Aspect RatiosAdaptive, 16:9, 9:16, 1:1, 4:3, 3:4 (default 16:9)
GenerationSingle pass - no stitching or upscale step

Prompt & Inputs

PromptRequired, up to 2,500 characters
Reference ImageNot accepted
Reference VideoNot accepted
Reference AudioNot accepted
DocumentsNot accepted

On SharkFoto

ToolAI Video - Text to Video
DeveloperAlibaba
ModesText to video only
CreditsScale with both resolution and clip length
Other ModesUse Wan 2.6 for image-to-video and video-to-video
In practice

Use cases

Vertical Social Cuts

Generate 9:16 clips sized for Reels, Shorts and TikTok without cropping a landscape render. Short durations keep the cost of testing several hooks low, and the same prompt can be re-run at 1:1 or 16:9 for the other placements.

Product and Marketing B-Roll

Fill the gaps in an edit - an establishing shot, a texture close-up, an ambient transition - without a shoot. Ask for the exact number of seconds the timeline needs rather than trimming a longer clip down.

Concept Previz

Test whether an idea reads on screen before anyone books a camera. Iterate the wording at 480P where a pass is cheap, then render the version that works at 1080P.

Article and Landing Page Visuals

A few seconds of motion at the top of a page or post does more than a static hero image, and 4:3 or 3:4 suits editorial layouts that a 16:9 clip would not sit in cleanly.

Multi-Format Ad Variants

One written concept, rendered into landscape, square and vertical versions, so a campaign ships in every placement without re-framing work in an editor.

Generational leap

Wan 2.6 vs Wan 3.0 on SharkFoto

Both Wan models are available here. This is what actually differs between them in the generator, not on paper.

FeatureWan 2.6Wan 3.0NEW
Clip lengthFixed 5s, 10s or 15sAny whole second, 1-15s
Resolution720p, 1080p480P, 720P, 1080P
Aspect ratios16:9, 9:16Adaptive + 16:9, 9:16, 1:1, 4:3, 3:4
Prompt limit2,000 characters2,500 characters
Draft tierNone below 720p480P
Generation modesText, image and video to videoText to video only
Credit modelPreset duration stepsCharged per second
Honest look

Current limitations

Text Prompts Only

Wan 3.0 on SharkFoto accepts a written prompt and nothing else. There is no way to supply a reference image, a reference video, an audio track or a document. Work that has to start from an existing asset belongs on Wan 2.6 or another image-to-video model.

Fifteen Seconds per Clip

A single generation tops out at 15 seconds. Longer pieces have to be assembled from multiple clips in an editor, and continuity between separate generations is not guaranteed.

1080P Costs Four Times 480P

Credits scale sharply with resolution: a 1080P clip costs four times the same-length 480P clip, and 720P costs twice as much. Iterating at 1080P gets expensive quickly - settle the prompt at 480P first.

No Editing or Continuation Modes

The model generates a new clip each time. It cannot extend an existing video, restyle footage, or edit a previous result - every run starts again from the prompt.

FAQ

Frequently asked questions

What is Wan 3.0?
Wan 3.0 is the latest generation of Alibaba's Wan video generation models, released publicly in August 2026. It turns a written description into a finished video clip in a single pass, planning the shot itself and holding subject and motion consistent for the length of the clip.
What can Wan 3.0 do on SharkFoto?
It runs as a text-to-video model. You write a prompt of up to 2,500 characters, choose a clip length between 1 and 15 seconds, pick 480P, 720P or 1080P, and choose one of six aspect ratios including Adaptive. The model returns a single video.
Can I use an image, a video or a document as a reference?
Not with Wan 3.0 here - it takes a text prompt only. Alibaba's own platform exposes a wider set of inputs for this model, but the endpoint SharkFoto uses accepts the prompt, ratio, resolution and duration and nothing more. For image-to-video or video-to-video, use Wan 2.6.
Can Wan 3.0 generate a 30-second video?
Not on SharkFoto. A single generation is capped at 15 seconds. Longer sequences have to be built from several clips and joined together in an editor.
What does the Adaptive aspect ratio do?
Adaptive leaves the framing to the model rather than forcing one of the fixed shapes. It is useful when the prompt describes a shot and you have no delivery format in mind. When the clip has to fit a specific placement, choose the ratio explicitly.
How are credits charged?
Per second of video, multiplied by the resolution tier. 720P costs twice as much per second as 480P, and 1080P costs four times as much, so a short 480P test is a small fraction of a full-length 1080P render.
Should I use Wan 3.0 or Wan 2.6?
Use Wan 3.0 when you are starting from a written idea and want precise control over length and framing, or want to iterate cheaply at 480P. Use Wan 2.6 when you need to animate an existing image, transform existing footage, or work with audio and video references.

Try Wan 3.0

Describe the shot, set the length to the second, pick your framing, and let Wan 3.0 render it. Start at 480P to find the prompt, finish at 1080P.

Try Wan 3.0 now