API Free WeekHundreds of APIs are free to call this week.Browse free APIs
Start building
MODEL COLLECTIONALIBABA NEXT-GENERATION VIDEO FAMILY

Wan 3.0

Create production-ready video from text, first and last frames, mixed references, existing footage, and audio. Wan 3.0 supports clips up to 30 seconds with native sound, precise motion, and consistent subjects.

Explore the lineupGet API key

Explore the model lineup

Preview the output style, compare capabilities, then open any model for full specs, pricing, and a live Playground.

Wan 3.0 Video output preview
LIVEvideo
MODEL PREVIEW
Alibaba/01

Wan 3.0 Video

Alibaba Wan 3.0 Video generates videos up to 30 seconds with native audio from text, a first/last frame, or reference images, videos, and audio.

  • text-to-video
  • image-to-video
  • reference-to-video
  • video-to-video
PRICEFrom $0.5
MODEvideo
View model detailsEXPLORE
Wan 3.0 Text to Video output preview
LIVEvideo
MODEL PREVIEW
Alibaba/02

Wan 3.0 Text to Video

Wan 3.0 by Alibaba turns a text prompt into a video of up to 30 seconds at 30fps, with native dialogue, music, and sound effects.

  • text-to-video
PRICEFrom $0.5
MODEvideo
View model detailsEXPLORE
Wan 3.0 Image to Video output preview
LIVEvideo
MODEL PREVIEW
Alibaba/03

Wan 3.0 Image to Video

Wan 3.0 by Alibaba - animate still images into dynamic videos with AI. Transform photos into cinematic clips with natural motion, camera movement, and optional audio generation.

  • image-to-video
PRICEFrom $0.5
MODEvideo
View model detailsEXPLORE
Wan 3.0 Reference to Video output preview
LIVEvideo
MODEL PREVIEW
Alibaba/04

Wan 3.0 Reference to Video

Wan 3.0 by Alibaba generates up to 30-second videos with native audio from reference images, videos, and audio, keeping subjects, motion, and voice consistent with the references.

  • reference-to-video
PRICEFrom $0.5
MODEvideo
View model detailsEXPLORE
Wan 3.0 Video Edit output preview
LIVEvideo
MODEL PREVIEW
Alibaba/05

Wan 3.0 Video Edit

Wan 3.0 Video Edit by Alibaba restyles or modifies an input video from a text instruction, optionally guided by reference images and audio, keeping the original length.

  • video-to-video
PRICEFrom $1
MODEvideo
View model detailsEXPLORE
Wan 3.0 Video Extend output preview
LIVEvideo
MODEL PREVIEW
Alibaba/06

Wan 3.0 Video Extend

Wan 3.0 Video Extend by Alibaba continues an input video forward from a text instruction, optionally guided by reference images and audio, with native audio.

  • video-to-video
PRICEFrom $0.5
MODEvideo
View model detailsEXPLORE
Wan 3.0 Prime Video output preview
LIVEvideo
MODEL PREVIEW
Alibaba/07

Wan 3.0 Prime Video

Alibaba Wan 3.0 Prime Video generates videos up to 30 seconds with native audio from text, a first/last frame, or reference images, videos, and audio.

  • text-to-video
  • image-to-video
  • reference-to-video
  • video-to-video
PRICEFrom $0.7
MODEvideo
View model detailsEXPLORE
Wan 3.0 Prime Text to Video output preview
LIVEvideo
MODEL PREVIEW
Alibaba/08

Wan 3.0 Prime Text to Video

Wan 3.0 Prime by Alibaba turns a text prompt into a video of up to 30 seconds at 30fps, with native dialogue, music, and sound effects.

  • text-to-video
PRICEFrom $0.7
MODEvideo
View model detailsEXPLORE
Wan 3.0 Prime Image to Video output preview
LIVEvideo
MODEL PREVIEW
Alibaba/09

Wan 3.0 Prime Image to Video

Wan 3.0 Prime by Alibaba - animate still images into dynamic videos with AI. Transform photos into cinematic clips with natural motion, camera movement, and optional audio generation.

  • image-to-video
PRICEFrom $0.7
MODEvideo
View model detailsEXPLORE
Wan 3.0 Prime Reference to Video output preview
LIVEvideo
MODEL PREVIEW
Alibaba/10

Wan 3.0 Prime Reference to Video

Wan 3.0 Prime by Alibaba generates up to 30-second videos with native audio from reference images, videos, and audio, keeping subjects, motion, and voice consistent with the references.

  • reference-to-video
PRICEFrom $0.7
MODEvideo
View model detailsEXPLORE
Wan 3.0 Prime Video Edit output preview
LIVEvideo
MODEL PREVIEW
Alibaba/11

Wan 3.0 Prime Video Edit

Wan 3.0 Prime Video Edit by Alibaba restyles or modifies an input video from a text instruction, optionally guided by reference images and audio, keeping the original length.

  • video-to-video
PRICEFrom $1.4
MODEvideo
View model detailsEXPLORE
Wan 3.0 Prime Video Extend output preview
LIVEvideo
MODEL PREVIEW
Alibaba/12

Wan 3.0 Prime Video Extend

Wan 3.0 Prime Video Extend by Alibaba continues an input video forward from a text instruction, optionally guided by reference images and audio, with native audio.

  • video-to-video
PRICEFrom $0.7
MODEvideo
View model detailsEXPLORE

Choose the right Wan 3.0 workflow

Start with the unified endpoint or choose a focused endpoint for a simpler production contract. Prime variants trade a higher rate for premium output quality.

WorkflowInputBest forKey capabilities
Wan 3.0 VideoText, first/last frame, images, video, or audioTeams that want one endpoint for the full creative workflowUp to 30 seconds, 480p/720p/1080p, native audio, mixed references
Text to VideoPromptConcept films, ads, cinematic shots, and social video2–30 second duration, multiple aspect ratios, synchronized audio
Image to VideoFirst frame and optional last frameAnimating product photography, storyboards, and key artNatural motion, camera movement, frame interpolation, audio generation
Reference to VideoUp to 10 images, 5 videos, and 5 audio clipsCharacter consistency, brand films, and multi-shot creative directionMixed references, subject consistency, motion and voice guidance
Video EditExisting video plus a text instructionRestyling footage, changing lighting, and controlled local editsPreserves source timing while applying instruction-led changes
Video ExtendExisting video plus optional referencesContinuing a shot while preserving its visual and audio languageForward continuation with native sound and reference guidance

COMPLETE VIDEO WORKFLOW

One model family from first frame to final cut

Wan 3.0 is Alibaba's next-generation video model family for turning prompts and visual references into longer, more controllable clips. It combines text-to-video, first-and-last-frame animation, multi-reference generation, video editing, and video extension behind a consistent SandBase API.

Built for longer, controlled generation

  • Up to 30 seconds: Generate complete shots without stitching together a sequence of short clips.
  • Native audiovisual output: Generate dialogue, music, ambience, and sound effects alongside the video.
  • Mixed-reference control: Guide identity, products, movement, voice, and visual language with images, video, and audio.
  • First and last frames: Define how a shot begins and ends, then let Wan 3.0 create the transition.
  • Production resolutions: Choose 480p, 720p, or 1080p and common landscape, square, and portrait aspect ratios.
  • Editing and extension: Transform existing footage or continue a shot without starting over.

Standard or Prime

Use Wan 3.0 for iteration, high-volume production, and cost-sensitive workloads. Choose Wan 3.0 Prime when final-delivery quality is the priority. Both tiers support the same core workflows, making it easy to prototype on Standard and promote selected shots to Prime.

Where Wan 3.0 fits

The family is well suited to advertising, short drama, music video, product visualization, branded character content, education, and social campaigns. The unified endpoint is the fastest way to start; focused endpoints are useful when an application exposes only one input mode.

How Wan 3.0 compares

Choose based on duration, reference control, audiovisual output, and the quality tier your production needs.

ModelStrengthsBest forTrade-offs
Wan 3.0Up to 30 seconds, native audio, mixed references, editing, and extensionLonger narrative shots and end-to-end creative pipelinesComplex multi-reference requests benefit from careful prompt and asset preparation
Wan 3.0 PrimeThe same broad input modes with a premium quality tierHero shots, client delivery, and final campaign assetsHigher per-second price than the Standard tier
Seedance 2.5Strong cinematic motion, prompt adherence, and polished short-form outputCommercial clips and highly directed visual sequencesCompare reference handling and native duration for each project
Wan 2.7Established short-form text-to-video and image-to-video workflowsExisting Wan pipelines and focused generation jobsLess flexible than Wan 3.0 for long clips and mixed-reference production

Start using Wan 3.0 on SandBase

Use one SandBase API key to call the unified Wan 3.0 endpoint, then poll the returned prediction until the video is ready.

  1. 01

    Create an API key

    Create a key in the SandBase Console and store it in an environment variable. Never commit production credentials to source control.

    TERMINAL

    export SANDBASE_API_KEY="YOUR_SANDBASE_API_KEY"
  2. 02

    Submit a video generation

    Call the unified run endpoint with the Wan 3.0 model and your generation parameters.

    TERMINAL

    curl https://api.sandbase.ai/v1/run \
  -H "Authorization: Bearer $SANDBASE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"alibaba/wan/3.0/video","input":{"prompt":"A cinematic tracking shot follows a sailboat across a glowing sea at sunset, natural wind and ocean ambience","duration":5,"resolution":"720p","aspect_ratio":"16:9","generate_audio":true}}'
  3. 03

    Add visual references

    Use image and end_image for a first/last-frame shot, or provide images, videos, and audios for a mixed-reference production. Open the model Playground to validate the request before integrating it.

Questions, answered

What is Wan 3.0?

Wan 3.0 is Alibaba's next-generation video model family. It can create video from text, images, video, and audio, generate native sound, edit existing footage, and extend a shot.

How long can a Wan 3.0 video be?

Wan 3.0 supports generated clips from 2 to 30 seconds. You can choose an exact duration or use automatic duration when supported by the selected endpoint.

What resolutions and aspect ratios are supported?

The SandBase Wan 3.0 endpoints support 480p, 720p, and 1080p output, with landscape, portrait, square, and cinematic aspect ratios including 16:9, 9:16, 1:1, 4:3, 3:4, and 21:9.

What references can I provide?

The unified reference workflow accepts up to 10 images, 5 short video clips, and 5 short audio clips. First-frame and last-frame inputs are also available for controlled image-to-video generation.

What is the difference between Wan 3.0 and Wan 3.0 Prime?

Both tiers cover the same core workflows. Standard is a strong default for iteration and scale; Prime is intended for premium final output and has a higher per-second price.

Can Wan 3.0 edit or extend an existing video?

Yes. Use Video Edit to restyle or modify footage from an instruction, and Video Extend to continue a clip while preserving its visual and audio direction.

READY / SET / RUN

Build with Wan 3.0.

One key, consistent APIs, and usage-based pricing across the entire lineup.

Get your API keyRead the docs