3.0

Wan 3.0 AI Video Generator. Bring your ideas to life.

Turn text, images, and references into videos with synchronized audio. Create clips up to 30 seconds in 1080p, right in your browser.

Get started with 10 free credits

A 20-second 21:9 ultra-widescreen plush universe…

Create

One continuous handheld shot follows an old man…

Create

Generated with reference to image1 and image2.…

Create

Introducing Wan 3.0

Meet the Wan 3.0 model. One complete story.

Alibaba Cloud's Wan 3.0 model works with text, images, and reference media. XRMM provides a separate browser-based workspace where you choose your inputs, generate, and download.

Model documentation (opens in a new tab)Explore Wan 3.0

Text to video

Describe a subject, scene, movement, and sound. Start from an idea without uploading a starting image.

Create from text

Wan 3.0 image to video

Start with a still image and describe how it should move. Use a product photo, portrait, or scene as your visual starting point.

Animate an image

Reference to video

Combine reference media to guide a scene. Check upload limits and the credit estimate before generating.

Create with references
01

Omni-Creation

Generate with up to 20 reference assets, including creative material parsed from documents and webpages.

02

Native 30s Duration

Tell longer, more complete stories with native 30-second generation and intelligent control over duration.

03

Pixel-Perfect Consistency

Reproduce reference details with greater precision for more predictable, production-ready results.

04

Immersive Experience

Bring richer realism, visual texture, and sound design together for a more convincing audiovisual experience.

05

Precision Video Editing

Follow edit instructions more accurately, using prompts and references to refine video with greater creative efficiency.

Portrait Customization Wan 3.0 feature example
06

Portrait Customization

Shape distinctive faces through bone structure, eye details, and subtle features for more authentic characters.

Wan series

Pick the Wan model for the job.

Compare length, credits, output, and what you can upload. Start with 3.0 for a finished scene, or open an earlier model for a tighter workflow.

3.0

Current

Wan 3.0

Complete scenes with picture and sound in one pass.

Duration

2–30s; reference video shares the 30s limit

Output

Native 1080P

Credits

Standard: 2/3/6 credits/s · Prime: 2/4/8 credits/s (480p/720p/1080p). Reference video seconds also count.

Audio

Built in

Uploads

10 images, 5 videos, 5 audio

Formats

JPG, PNG, BMP, WEBP · MP4, MOV · MP3, WAV

2.7

Wan 2.7

Pro controls, subject lock, and video edit.

Duration

2–15s

Output

720P–1080P

Credits

1080P 3/s · 720P 2/s

Audio

Optional

Uploads

2 frames, 1 video, 3 edit images, 1 audio

Formats

JPG, PNG, WEBP · MP4, WebM, MOV · MP3, WAV

2.6

Wan 2.6

Multi-shot stories and video-to-video.

Duration

15s

Output

1080P 24fps

Credits

5s / 10s / 15s · 1080P 15 / 30 / 45 · 720P 9 / 18 / 30

Audio

Synced

Uploads

1 image, 2 videos, 1 audio

Formats

JPG, PNG, WEBP · MP4 · MP3, WAV

2.5

Wan 2.5

Talking clips with lip-sync and voice drive.

Duration

5–10s

Output

480P–1080P

Credits

5s / 10s · 1080P 15 / 30 · 720P 9 / 18 · 480P 5 / 10

Audio

Voice + lip-sync

Uploads

1 image, 1 audio

Formats

JPG, PNG, WEBP · MP3, WAV

2.2

Wan 2.2

Fast cinematic drafts from a prompt or still.

Duration

5s

Output

480P–720P

Credits

1080P 8 · 720P 4 · 480P 2 per clip

Audio

Upload

Uploads

1 image, 1 audio

Formats

JPG, PNG, WEBP · MP3, WAV

Wan 3.0 capabilities

Create from any direction. Finish in one model.

Move between references, images, prompts, and existing footage while keeping the creative language of your project intact.

Reference to Video

Cast referenced characters into new scenes, supporting multi-person and human-object interactions with consistent appearance and voice.

Image to Video

Generate videos up to 30 seconds with auto scene splitting for more stable native audio-visual videos.

Text to Video

Cinematic precision with sophisticated motion and highly reliable prompt execution.

Video Editing

Multi-dimensional instruction-based video editing for re-creating visuals, storylines, and environments.

Three simple steps

How to Create Video with Wan 3.0

Move from an idea to a finished Wan 3.0 video without a separate editing workflow.

Describe Your Video

Start with a clear prompt. Define the subject, action, setting, camera direction, visual mood, and the sound you want to hear.

Add References and Settings

Upload an image when you need stronger visual direction, then choose the duration, aspect ratio, and output settings for your scene.

Generate and Download

Let Wan 3.0 create the picture and synchronized audio together. Review the result, refine your direction, or download the final video.

Created with Wan 3.0

Wan 3.0. One model. Endless directions.

Explore cinematic scenes, product stories, stylized characters, and visual experiments generated with Wan 3.0.

Create your video

Simple, one-time pricing

Create more with Wan 3.0.

Choose a credit pack once. Your credits stay available until you use them.

Starter

$9.9
One-time payment · 100 credits

What's included

  • $0.0990 per credit
  • 100 credits included
  • create HD text-to-video or image-to-video clips with natural native audio
  • 720p export, No watermark download
  • Commercial use license
  • Standard queue speed
  • Email support

Basic

$29.9
One-time payment · 330 credits

What's included

  • $0.0906 per credit
  • 330 credits included
  • faster HD generation for daily content
  • Text to Video & Image to Video with native audio
  • 1080p export, No watermark download
  • Commercial use license
  • Priority queue speed
  • Priority support (email)

Professional

$99.9
One-time payment · 1250 credits

What's included

  • $0.0799 per credit
  • 1250 credits included
  • high-volume, professional delivery and teams
  • Text to Video & Image to Video with native audio
  • 1080p export, No watermark download
  • Commercial use license
  • Fastest queue + up to 10 concurrent jobs
  • Full effects pack + early access to new features
  • 24/7 priority support
  • Bulk processing
  • API access (coming soon)
7‑Day Refund
Money-back guarantee
Secure Payment
Powered by Stripe
24/7 Support
Always here to help

Choose one-time credits • Flexible billing options

Choose one-timeCredits never expireSecure paymentsEmail support support@xrmm.com

Practical answers

Wan 3.0 questions, answered clearly.

Learn what Wan 3.0 does, how to start with 10 credits, and what to check before creating your first video.

What is Wan 3.0?
Wan 3.0 is an AI video generation model for creating clips from text, images, and reference media. On XRMM, you can use it in your browser, choose Standard or Prime, and review your settings and credit estimate before generating.Open the Wan 3.0 generator
How can I get started with Wan 3.0 AI?
Open the generator and sign in. Choose text, image, or reference mode, add your prompt and any required media, then select resolution and duration. Check the credit estimate before starting your task.Create your first video
Is Wan 3.0 free to use?
XRMM provides 10 complimentary credits to get started. These are credits, not a fixed number of free videos. The model, resolution, output duration, and reference video duration determine the cost. Choose settings within your balance or purchase a one-time credit pack.Read about free credits
How much does a Wan 3.0 video cost?
Wan 3.0 uses credits per billable second, with different rates for Standard and Prime at each resolution. Reference video duration also counts toward billing. The pricing page shows the rates, pack prices, and worked examples.Compare rates and video costs
How long can a Wan 3.0 video be?
The generator supports output durations from 2 to 30 seconds. In reference mode, reference video duration and output duration share a 30-second limit, so adding a reference video reduces the available output length. Check the current settings before generating.
How many references can I combine in one Wan 3.0 project?
Wan 3.0 can work with up to 20 reference assets. Use each source intentionally—for example, one for the character, another for the setting, and a separate reference for motion, styling, or sound direction.
Should I begin with text or an image?
Start with text when you want broad creative freedom. Begin with an image when composition, product appearance, character identity, or art direction needs to stay closer to a defined visual source.
How should I plan a full 30-second generation?
Outline a small sequence of visual beats before writing the prompt: establish the scene, develop the action, introduce the key change, and define the final frame. Clear timing cues help the model distribute motion across the full clip.
How can I reduce visual drift from my references?
Use clean reference assets, name the details that must remain unchanged, and avoid giving competing style instructions. When iterating, keep the successful parts of the prompt stable and change only one weak element at a time.
Can I direct sound separately from the picture?
Yes. Separate visual direction from audio direction inside the prompt. Describe dialogue, ambience, music, and important sound cues explicitly, including when each sound should enter or respond to the on-screen action.
What should I review before downloading a Wan 3.0 video?
Check subject consistency, hand and object interactions, camera movement, framing, text accuracy, and audio timing. If one area needs work, revise that instruction specifically instead of replacing the entire prompt.
Can Wan 3.0 outputs be used in commercial projects?
Commercial usage depends on the credit plan and the rights attached to your uploaded materials. Confirm that you have permission to use every reference asset and review the applicable product terms before publishing client or advertising work.

Create Your First Video with Wan 3.0

Turn a prompt or image into a cinematic 1080P video with synchronized audio in one streamlined Wan 3.0 workflow.

Create with Wan 3.0