QwenQwen Video generation

Wan 3.0 Video

Turn a shot brief into a reviewable video take.

Wan 3.0 Video is the standard Wan 3.0 video route in TokenHot's catalog. It is described as an early-access text-and-image-to-video path with video output, configurable resolution tiers and a catalog note about synchronized audio. Use it for short product shots, story beats and reference-led motion studies while treating availability and parameters as versioned checks.

  • Text + image input
  • Video output
  • Early access route

The catalog currently labels this route early access. Check the full pricing table and API docs for duration, resolution, audio and reference-input support before promising a production workflow; the page does not treat the illustration or a prompt as proof of generated quality.

USD /s

Pricing

GroupTierPrice
default480P$0.045
default720P$0.09
default1080P$0.18
default
480P$0.045
720P$0.09
1080P$0.18

Public list prices are shown in USD. Final charges may vary by account group and usage tier.

Overview

What is Wan 3.0 Video?

A useful video request names the subject, action, camera movement, framing and ending state. A reference image can anchor the opening frame, but it does not guarantee identity or continuity through the whole clip.

Wan 3.0 Video is distinct from the prime route and from other video families. Keep the model ID, resolution and duration in your job record so a later rerun is explainable.

Shot direction

Write visible events

Separate subject movement, camera movement and the final frame. Avoid relying on mood words to carry the shot.

Reference control

Define what must persist

Name the composition and identity details the reference should inform, then review drift in the resulting take.

Review loop

Judge timing before polish

Start with a short, single-purpose shot. Check motion, continuity, framing and audio before assembling a larger edit.

From brief to deliverable

Put Wan 3.0 Video to work

Start with a concrete task and decide what a useful result looks like.

Creative and marketing teams

Create a product reveal study

Use a product reference and a concise shot brief to explore camera movement, lighting and the final hold. Treat the output as a storyboard-quality take until reviewed.

Tools you need
A video generation client, asset storage and a human review pass.
Keep in mind
Early-access behavior and identity consistency need direct review; do not promise a final campaign asset from one generation.
View a task brief

Use the supplied a product image, shot duration, resolution target and a description of the reveal. to plan create a product reveal study. Return a reviewable video take with a shot checklist., state the limits of the evidence and identify the checks required before accepting the result. Do not claim that an output or interaction was verified without direct evidence.

Directors and content teams

Prototype one narrative beat

Describe one character action and one camera move rather than a full scene. Use the result to decide whether the beat deserves editorial development.

Tools you need
A generation queue, a video player and an editing tool for assembly.
Keep in mind
The route does not replace continuity planning, audio editing or final color and legal review.
View a task brief

Use the supplied a beat outline, subject details, location cues and the intended final frame. to plan prototype one narrative beat. Return a short motion study and notes on continuity., state the limits of the evidence and identify the checks required before accepting the result. Do not claim that an output or interaction was verified without direct evidence.

Design and social teams

Animate a reference composition

Start from a still frame and ask for a controlled motion change, such as a slow push-in or a single gesture. Compare the result against the reference before adding complexity.

Tools you need
Reference upload, output storage and a human comparison pass.
Keep in mind
Reference inputs guide the request but cannot guarantee exact text, hands, faces or brand details.
View a task brief

Use the supplied a reference image, motion constraint, aspect ratio and delivery target. to plan animate a reference composition. Return a motion variation with a pass/fail review sheet., state the limits of the evidence and identify the checks required before accepting the result. Do not claim that an output or interaction was verified without direct evidence.

Capabilities in context

Key features

FeatureWhat you getWhy it matters
Input modesText and image input are listedUse a shot brief or a first-frame reference, and validate the accepted media shape before submitting.
Video outputThe route is cataloged as a video modelTreat each result as an asynchronous job and retain the prompt, model ID and output metadata.
Resolution tiersPricing expression exposes 480P, 720P and 1080P labelsResolution affects cost and review expectations. Keep it explicit in the job request and acceptance sheet.
Duration and audioThe catalog description mentions clips up to roughly 30 seconds and synchronized audioThese are catalog descriptions, not a guarantee for every route configuration. Verify duration and audio support in the API response.
Early accessThe catalog description identifies early accessKeep a fallback or manual review path for launch-critical work and record any behavior changes.

This page uses TokenHot catalog metadata and its current billing expression. Wan 3.0 Video is marked early access; verify job parameters, availability and media constraints before production use.

Audience & fit

Where Wan 3.0 Video fits

Creative teams exploring short-form ideas

Use focused shots to explore motion and composition before investing in a full edit or campaign asset.

Product and social teams

Provide a clear reference and destination constraint. Keep human review for brand identity, typography and continuity.

Developers building media workflows

Treat generation as an asynchronous, versioned job. Store prompts, parameters, status and output metadata for reruns.

Cost context

Plan the cost of a useful result

Price by configuration

Video cost changes with resolution and duration. Use the full dynamic pricing table and keep the job parameters beside the estimate.

Budget failed takes

A visually unusable or continuity-breaking take still consumes a generation. Include review and rerun rates in the project estimate.

Keep media processing separate

Reference preparation, storage, transcoding and editing add work outside the model call. Track those costs independently.

Use the current pricing table for available rates and account groups. Model IDs and prices come from TokenHot’s catalog; provider pricing and subscriptions are separate.

API

Code examples

curl --location --request POST 'https://api.tokenhot.ai/v1/video/generations' \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data-raw '{
    "model": "wan3.0-video",
    "input": {
        "prompt": "A miniature city built from cardboard and bottle caps comes to life at night. A cardboard train slowly passes by, with small lights dotted along it, illuminating the path ahead."
    },
    "parameters": {
        "resolution": "720P",
        "ratio": "16:9",
        "duration": 5
    }
}'

Qwen

Related models

QwenQwenWan3.0 Video Prime

The speed-first variant of Alibaba's Wan 3.0 line (early access). It shortens generation latency while retaining high-quality video output, making it a good fit for low-latency previews, batch drafts and fast-iterating short-video workflows. If ultimate visual control matters more than speed, the standard version is a better choice.

$0.0675/sQwenQwenHappyhorse 1.1 T2V

Built for text-to-video, with a stronger audio-native workflow than HappyHorse 1.0, generating dialogue, sound effects, and background music in one pass. Compared with Veo, Kling, and Runway, its differentiator is synchronized audiovisual generation and character-consistent workflows for short drama, ads, e-commerce, and brand marketing.

$0.135/sQwenQwenHappyhorse 1.1 I2V

Built for image-to-video, with a stronger audio-native workflow than HappyHorse 1.0, generating dialogue, sound effects, and background music in one pass. Compared with Veo, Kling, and Runway, its differentiator is synchronized audiovisual generation and character-consistent workflows for short drama, ads, e-commerce, and brand marketing.

$0.135/sQwenQwenHappyhorse 1.0 T2V

Built for text-to-video, with a stronger audio-native workflow than 早期视频模型, generating dialogue, sound effects, and background music in one pass. Compared with Veo, Kling, and Runway, its differentiator is synchronized audiovisual generation and character-consistent workflows for short drama, ads, e-commerce, and brand marketing.

$0.135/s

FAQ

Frequently asked questions

What is Wan 3.0 Video best suited for?

Wan 3.0 Video is suited to the input, output, and capability types listed on this page. Test production workloads before choosing it for a critical system.

How is Wan 3.0 Video priced?

Pricing is shown in the pricing section above. Actual cost depends on usage volume, account group, and the request payload.

How can I call Wan 3.0 Video?

Use the model ID shown above with one of the supported API protocols. Code examples are provided when usage data is available.

Is Wan 3.0 Video generally available?

The current TokenHot catalog description labels it early access. Check the live route and API response before making a production availability promise.

Can it generate audio with the video?

The catalog description mentions synchronized audio, but exact support depends on the endpoint configuration. Verify the returned job and parameters for your route.

How should I write a video prompt?

Name the subject, visible action, camera movement, framing, duration and final frame. Keep the first test to one shot with one primary motion.

Is this the same as Wan 3.0 Video Prime?

No. They are separate TokenHot model IDs with potentially different controls and pricing. Use the exact ID shown in your request.

Get started

Build with Wan 3.0 Video

Use in Console