AnydocAILOCAL + CLOUD
Open workspace
MESSY DOCUMENTS → AI-READY OUTPUT

See the result.
Then decide to pay.

Turn PDF, Word, PowerPoint, Excel, EPUB, and scanned documents into clean Markdown, JSON, and RAG-ready chunks.

Native parser Cloud OCR with consent 21+ formats
LOCAL-FIRST CONVERTER SCANNED PDFS USE SECURE CLOUD OCR
Drop a document here

or click to browse · up to 50 MB

PDFDOCXPPTXXLSX+ 17
SEE REAL OUTPUT
21 FORMATSWORDPOWERPOINTEXCELPDFOPENDOCUMENTRTFEPUBCSV
01 / WORKSPACE

From a folder of files
to model-ready data.

One pipeline for every document your users upload. Preserve structure, normalize output, and stop maintaining format-specific parsers.

A/
BATCH / 0048

Product research

OUTPUT RAG chunksSIZE800 tokensOVERLAP80
q3-board-deck.pptx42 slides
READY
research-notes.docx18 pages
READY
market-data.xlsx7 sheets
PARSING
scanned-contract.pdf12 pages
QUEUED
3,281 CHUNKS17.4 MB OUTPUT3 / 4 COMPLETE
01

Mixed batches

Process office files, PDFs, spreadsheets, ebooks, and legacy formats together.

02

Reviewable output

Preview, copy, or download clean Markdown before it enters another system.

03

Automatic OCR

Scanned PDFs fall back to hosted OCR while ordinary documents stay on the fast native path.

02 / PIPELINE

Three steps. Zero parser glue.

01

Upload

Send one file or a mixed batch from the dashboard or API.

FILES RECEIVED
02

Normalize

Detect formats, preserve structure, run OCR, and create chunks.

STRUCTURE PRESERVED
03

Export

Copy the result or download a Markdown file for your next workflow.

MARKDOWN READY
03 / DEVELOPER WORKFLOW

Built for the ingestion path.

Normalize mixed documents into Markdown, structured JSON, or deterministic chunks, then receive a signed webhook when the conversion finishes.

  • Bearer API keys
  • OCR fallback for scanned PDFs
  • Markdown, JSON, and RAG chunks
Read the API reference
REQUESTRESPONSE200 OK
curl https://anydocai.com/api/v1/documents/convert \
  -H "Authorization: Bearer $ANYDOCAI_API_KEY" \
  -F "file=@report.pdf" \
  -F "output=chunks" \
  -F "ocr=auto"
{
  "object": "document.conversion",
  "status": "completed",
  "output": "chunks",
  "ocrUsed": false
}
LOCAL MODE

Digital files stay in your browser.

Compatible documents use WebAssembly locally. We show the active processing path before work begins.

PRIVATE BY DEFAULT
CLOUD OCR

Scanned pages need hosted processing.

We ask for consent before a scanned file is securely uploaded for OCR and delete the source after processing.

CONSENT REQUIRED
04 / SIMPLE PRICING

See the output.
Pay only when it works.

Start with a free preview. Buy one result or a page pack without committing to a subscription.

Free preview

See your own result first.

$0
  • First 10 pages
  • No account required
  • Real output preview
Upload a document

Single File

Unlock the file you previewed.

$5 once
  • Up to 200 pages
  • Markdown + JSON + chunks
  • 7-day access
Preview first
BEST FIRST PURCHASE

Starter Pack

Best for several documents.

$9 once
  • 500 processed pages
  • Up to 20 files
  • Valid for 90 days
See pricing
05 / QUESTIONS

Built to be
straightforward.

Does AnydocAI upload files?+

Compatible digital files are converted locally. Scanned documents require cloud OCR, and we ask for consent before uploading them.

What happens with scanned PDFs?+

We explain the cloud processing path before upload. Source files are deleted after processing according to the published retention policy.

Which output can I generate?+

The API generates Markdown, structured JSON, or RAG-ready chunks.

Can I process several files?+

Yes. Signed-in users can add mixed document formats to the cloud workspace queue.

What is recorded for cloud jobs?+

The workspace records bounded task metadata and a short result preview; the response returns the full Markdown to your session.