Best for · Document-to-video

Document to Video AI: 9 Formats, 1 Model Accepts Them

Wan 3.0 is the only major video model that accepts documents directly, taking doc, xls, ppt, pdf, txt, key, pages, numbers and md files plus webpage URLs.

Why document-to-video is its own niche

Most video models accept text and image inputs only. Slide-deck and PDF-to-video workflows have historically required a manual pipeline: extract text, summarize, generate a script, generate a video. Wan 3.0 collapses this into one call.

Accepted formats

FormatAccepted by Wan 3.0Accepted by Seedance 2.0Accepted by MiniMax H3
PDFYesNoNo
PowerPoint (.ppt)YesNoNo
Word (.doc)YesNoNo
Excel (.xls)YesNoNo
Plain text (.txt)YesYes (as prompt)Yes (as prompt)
Apple Keynote (.key)YesNoNo
Apple Pages (.pages)YesNoNo
Apple Numbers (.numbers)YesNoNo
Markdown (.md)YesYes (as prompt)Yes (as prompt)
Webpage URLYesNoNo

How to use document-to-video

  • POST to /v1/ula/generate-video with model "alibaba/wan3.0-video".
  • Pass document_reference as the public URL or signed upload path.
  • Add a prompt that names the desired narrative — bullet-by-bullet, executive summary, or Q&A.
  • Set duration to cover the source length — 60 seconds for a typical 12-page deck.

Ready to test the workflow?

Create account & add credits

When to use the seedance pipeline instead

Seedance 2.0 and MiniMax H3 do not read documents natively. If you need to stay on those gateways, pre-process the document into a prompt manually. Expect a 10-30 minute setup cost that Wan 3.0 eliminates.

Frequently asked questions

What is the best PDF to video AI?

Wan 3.0 is the only major video model that accepts PDF files directly. Pass the PDF URL as document_reference in a POST to /v1/ula/generate-video.

Can I turn a PowerPoint into a video?

Yes — pass the .ppt URL as document_reference to Wan 3.0. The model reads slide content and produces a narrated walkthrough at the requested duration and resolution.

Where is the canonical Wan 3.0 page on OneInfer?

The canonical model page with current pricing, providers, and capabilities is /models/alibaba/wan3.0-video. This page is a focused facet of that entity, not a replacement for it.

How should I treat benchmark or price claims?

Check each claim's provenance label and observed date. Vendor-reported and independently verified numbers are shown as separate evidence classes on this hub.

How do I call Wan 3.0 from the API?

POST to /v1/ula/generate-video with model "alibaba/wan3.0-video", a valid API key, and a prompt plus aspect_ratio and duration. Add first_frame, last_frame, or document_reference only when you need them.

Put Wan 3.0 to work

Fund a controlled evaluation, send a reference frame or document, and measure quality and cost on your own workload.