Wan 3.0

Turn documents and web pages into complete 30-second videos.

Professional video for every use case

Document-to-video briefings.

Transforms reports, decks, and spreadsheets into coherent narrated visual sequences.

30-second brand spots.

Fits a complete campaign arc, synchronized sound, and continuous camera movement into one generation.

Reference-led product demos.

Keeps packaging, materials, logos, and presenter details stable across product-focused shots.

Vertical creator videos.

Creates portrait clips with direct-to-camera speech, natural gestures, and matched audio.

UI and data explainers.

Animates interfaces, charts, and structured information while preserving a polished visual hierarchy.

Native 30-second scenes

Wan 3.0 can generate up to 30 seconds in a single pass, leaving room for complete narrative beats and continuous camera movement. Smart duration can match clip length to the prompt, while extension supports longer sequences.

  • Up to 30 seconds per generation
  • Smart duration recommendations
  • Continuous one-take camera movement
  • Video extension for longer narratives
  • Vertical and landscape storytelling

Mixed references in one creative brief

Guide a generation with text, images, video, audio, documents, and public web pages. Wan 3.0 uses these materials together to control characters, products, locations, voices, style, and story details.

  • Up to 10 reference images
  • Up to 5 reference videos
  • Up to 5 reference audio tracks
  • Support for common document formats
  • Public webpage references
  • Position-based reference instructions

1080p video and sound in one pass

Wan 3.0 generates the visual track and audio together rather than treating sound as a separate finishing step. This supports dialogue, lip movement, ambience, music, and action that follow the same scene direction.

  • 480P, 720P, and 1080P output
  • Synchronized dialogue and lip movement
  • Ambient sound and music direction
  • Multilingual voice performances
  • Portrait, landscape, and adaptive framing

Stable details with targeted editing

Reference-to-video workflows preserve fine details across characters, products, spaces, voices, and visual styles. Editing controls can revise visuals, plot, or dialogue without rebuilding the entire concept from the beginning.

  • Consistent faces, clothing, and accessories
  • Stable product structure and materials
  • Preserved logos and prop details
  • Controlled spatial relationships
  • Consistent voices and visual styles
  • Visual, story, and dialogue revisions

How it works

Describe your story
1

Describe your story

Write the scene, action, camera direction, dialogue, and sound you want. Add images, video, audio, documents, or web references when you need tighter control over the result.

Choose length and quality
2

Choose length and quality

Set a duration of up to 30 seconds and choose the resolution that fits your workflow. Use a lower resolution for quick drafts, then move to 1080P for a polished final generation.

Generate and refine
3

Generate and refine

Generate the clip and review motion, subject consistency, framing, dialogue, and sound. Adjust your direction or references, then create another pass until the sequence matches your intent.

Pricing for Wan 3.0

Runs on credits — no per-model surcharges, no surprise billing.

170credits
per second of video
≈ 850 credits for a 5-second clip
Show pricing details
ResolutionCredits
  • 720pdefault170/ sec
  • 1080p340/ sec
  • 480p85/ sec

Credits work across every plan. See /pricing for credit packages.

Use Wan 3.0 via the API

Wan 3.0 is available through the BudgetPixel developer API — the same model the studio runs, supporting text-to-video, image-to-video, and reference-to-video. Pricing is metered in credits (per tier below), charged only on success, with an API key available on Premium plans and above.

EndpointPricingDocs
POST /v1/videos/wan-3.0-video60 credits/second at 480p, 120 credits/second at 720p, 240 credits/second at 1080pWan 3.0 docs
POST /v1/videos/wan-3.0-video-prime85 credits/second at 480p, 170 credits/second at 720p, 340 credits/second at 1080pWan 3.0 Prime docs
curl -X POST https://api.budgetpixel.com/v1/videos/wan-3.0-video \
  -H "Authorization: Bearer $BUDGETPIXEL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "your prompt here"}'

No code? Use Wan 3.0 from Claude or Cursor

Wan 3.0 is also available through the BudgetPixel MCP server. Connect your account to Claude, Cursor, or any MCP-enabled agent and generate with Wan 3.0 by asking in plain language — same credits, same pricing, no code required.

Set up the MCP server

Frequently asked questions

What is Wan 3.0?
Wan 3.0 is Alibaba's multimodal AI video generation model for creating clips from text and reference materials. It supports videos up to 30 seconds, resolutions up to 1080P, native audio generation, and document-to-video workflows.
How does Wan 3.0 work?
Wan 3.0 interprets your prompt and references as a unified creative brief covering subjects, action, camera movement, dialogue, and sound. It then generates the visual and audio tracks together, with smart duration controls that can match the clip length to the requested scene.
What inputs does Wan 3.0 accept?
Wan 3.0 accepts text, images, video, audio, documents, and public web pages. Its reference workflow can combine multiple media assets to guide character identity, products, locations, voices, motion, and visual style.
How long are Wan 3.0 videos, and what resolutions are supported?
Wan 3.0 supports clips up to 30 seconds in one generation. Available resolution tiers include 480P, 720P, and 1080P, with portrait and landscape framing supported for different publishing formats.
Does Wan 3.0 generate audio?
Yes. Wan 3.0 can generate dialogue, voices, ambience, music, and sound effects in the same pass as the video. This helps the audio follow on-screen action and supports synchronized lip movement for spoken performances.
Can Wan 3.0 turn documents into videos?
Yes. Wan 3.0 can use reports, presentations, spreadsheets, PDFs, text files, and other supported documents as source material. It can interpret their content and turn it into a structured briefing, explainer, training sequence, or promotional video.
How much does Wan 3.0 cost?
Generation cost is credit-based and scales mainly with the selected duration and resolution. Shorter or lower-resolution drafts use fewer credits, while longer 1080P videos require more.
How does Wan 3.0 compare to other AI video generators?
Wan 3.0 focuses on complete 30-second sequences, mixed-media reference control, document-to-video creation, and synchronized audiovisual output. It is especially relevant when a project needs longer narrative pacing or consistent characters, products, spaces, and voices.
Can I use Wan 3.0 videos commercially?
Wan 3.0 is a proprietary model accessed under service terms rather than an open-source license. Review the applicable provider and platform terms for commercial rights, restricted content, trademarks, reference media, and output ownership before publishing.
Can I edit a Wan 3.0 video after generating it?
Wan 3.0 supports workflows that revise visual content, plot details, and dialogue without restarting the entire concept. The exact editing controls available depend on how the model is accessed, so review the available options before generating.
Does Wan 3.0 have an API?
Yes — Wan 3.0 is available through the BudgetPixel developer API via 2 endpoints (Wan 3.0, Wan 3.0 Prime), supporting text-to-video, image-to-video, and reference-to-video. Generate an API key from the developer console (available on Premium plans and above) and see the full reference at docs.budgetpixel.com.
How much does the Wan 3.0 API cost?
API usage is billed in credits from the same balance your plan includes, at the standard metered rate: Wan 3.0: 60 credits/second at 480p, 120 credits/second at 720p, 240 credits/second at 1080p; Wan 3.0 Prime: 85 credits/second at 480p, 170 credits/second at 720p, 340 credits/second at 1080p. Failed generations are never charged. You can estimate any request's exact cost with POST /v1/cost before running it.
Can I use Wan 3.0 with Claude or other AI agents (MCP)?
Yes — Wan 3.0 is exposed through the BudgetPixel MCP server, so Claude, Cursor, and any MCP-compatible agent can generate with it from a plain-language request. Connect your account at budgetpixel.com/mcp; generations use the same credit balance and pricing as the web studio and the API.