Available now

Wan 3.0: text to video

By Alibaba

Alibaba's Wan 3.0 animates a starting image, follows up to 10 reference images or works from a prompt alone, with sound and clips up to 30 seconds long.

Model ID
Maker
Alibaba

Model id

Use this in every request
alibaba/wan-3.0/text-to-video

This model has other modes with their own ids: alibaba/wan-3.0/image-to-video, alibaba/wan-3.0/reference-to-video, wan-3-0.

Run it

Quote the input, start the generation with the quote, then poll the job. Set RAYWAKE_API_KEY first; see Authentication.

# 1. Price the exact input (quotes are valid for up to 10 minutes)
curl -s https://api.raywake.com/v1/quotes \
  -H "Authorization: Bearer $RAYWAKE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "alibaba/wan-3.0/text-to-video",
  "input": {
    "prompt": "A lighthouse keeper climbs a spiral staircase at night with a lantern while the wind howls outside. At the top she opens a small window and rain blows across her face."
  }
}'

# 2. Start the generation with the quote_id from step 1
curl -s https://api.raywake.com/v1/generate \
  -H "Authorization: Bearer $RAYWAKE_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{"model": "alibaba/wan-3.0/text-to-video", "quote_id": "QUOTE_ID", "input": {"prompt":"A lighthouse keeper climbs a spiral staircase at night with a lantern while the wind howls outside. At the top she opens a small window and rain blows across her face."}}'

# 3. Poll until the status is terminal
curl -s https://api.raywake.com/v1/jobs/JOB_ID \
  -H "Authorization: Bearer $RAYWAKE_API_KEY"

Input

Input fields for alibaba/wan-3.0/text-to-video
NameTypeDescription
promptrequiredstring
Text prompt for the generated video.
resolutionstring (enum)
Output video resolution tier.

Default: 1080p

One of: 480p, 720p, 1080p

aspect_ratiostring (enum)
Output aspect ratio, or adaptive selection.

Default: adaptive

One of: adaptive, 16:9, 4:3, 1:1, 3:4, 9:16

durationinteger
Output duration in seconds. Set to null for smart duration, which lets the model pick a length from the prompt and reference media.

Default: 5

audioboolean
Include generated audio.

Default: true

enable_prompt_expansionboolean
Enable intelligent prompt rewriting. Disabling it can save roughly 20-60 seconds of latency but is likely to degrade generation quality.

Default: true

enable_thinkingboolean
Enable enhanced reasoning before generation.

Default: false

seedinteger
Seed

Output

A succeeded job lists its results under outputs, stored on our side. See Jobs for every field.

Job excerpt
"outputs": [
  { "type": "video", "url": "https://…", "mime_type": "video/mp4", "thumbnail_url": "https://…" }
]

Pricing

  • 480p$0.0793/ second
  • 720p$0.1585/ second
  • 1080p$0.317/ second

The exact charge for your input comes from a quote. How charges settle

API docs

About this model

Overview

Wan 3.0 makes clips of 2 to 30 seconds at 480p, 720p or 1080p, with generated sound you can switch off. The adaptive aspect ratio takes the shape from your image and prompt; you can also pick 16:9, 1:1, 9:16 and others.

Start from a first image with an optional last image, give up to 10 reference images and call them by number in the prompt, or write a scene with no image at all. Prompt expansion and a thinking step can rework a short prompt before rendering.

What you can build

  • A product photo turned into a moving shot with ambient sound
  • Clips that start and end on frames you choose
  • Scenes that combine people and objects from several reference photos
  • Half-minute clips in one run, without stitching

Prompting tips

  • Add a last image when the clip has to land on a set frame, such as a logo or a final pose.
  • In the reference mode, name each picture by its position: "the woman in Image 1 holds the cup from Image 2".
  • Leave prompt expansion on for short prompts; turn it off when your prompt already says exactly what you want.

What to check

Compare the first and last frames with your sources. Watch for identity changes, drifting backgrounds and an abrupt ending; listen to generated sound when enabled.

Inputs

  • Prompt (required): Text prompt for the generated video.
  • Resolution (optional): Output video resolution tier. Options: 480p, 720p, 1080p.
  • Aspect Ratio (optional): Output aspect ratio, or adaptive selection. Options: adaptive, 16:9, 4:3, 1:1, 3:4, 9:16.
  • Duration (optional): Output duration in seconds. Set to null for smart duration, which lets the model pick a length from the prompt and reference media. From 2 to 30.
  • Audio (optional): Include generated audio.
  • Enable Prompt Expansion (optional): Enable intelligent prompt rewriting. Disabling it can save roughly 20-60 seconds of latency but is likely to degrade generation quality.
  • Enable Thinking (optional): Enable enhanced reasoning before generation.
  • Seed (optional) From 0 to 2147483647.

Output and limits

  • Length: 2 seconds to 30 seconds.
  • Resolutions: 480p, 720p, 1080p.
  • Aspect ratios: adaptive, 16:9, 4:3, 1:1, 3:4, 9:16.
  • Readable text inside the video, especially small text, is unreliable.
  • Scenes with several interacting characters can lose consistency.
  • Longer clips can show more motion drift or scene changes.
  • Low-quality inputs cause glitches: heavy compression, collages, screenshots and text overlays.
  • Some prompts or outputs can be refused by safety checks.

Price guide

  • 480p: $0.0793 / second
  • 720p: $0.1585 / second
  • 1080p: $0.317 / second
  • The total for your settings is rounded up to the next cent.