Question 1

Why can't I just let my model read the PDF? Claude and GPT have vision now.

Accepted Answer

Vision can get you the PDF. It doesn't get you the spreadsheet, the deck, the scanned contract, or the 60+ other formats sitting in a real corpus. Unstructured Transform reads all of them and hands your agent structured data back, so it's not limited to the handful of file types the model can open on its own.

Question 2

Why not just install your open-source and wrap it myself?

Accepted Answer

You can, and plenty of people start with the open-source library. What you're paying for is not building and babysitting the pipeline around it: strategy per file, scans versus clean text versus tables, growing format support, support at scale, and our latest parsing techniques as we release them.

Question 3

What do you actually build with this?

Accepted Answer

RAG pipelines: parse, chunk, and embed documents into a vector store. Semantic memory: turn a whole corpus into embeddings your agent can search back to later. Structured data extraction: pull specific fields into JSON against a schema you define, like vendor and total off an invoice, without becoming a document processing expert yourself.

Question 4

What does this actually cost me? My agent is already burning tokens.

Accepted Answer

15,000 pages a month free, then 3 cents a page after that. Since the parsing runs on Transform's side rather than in your model's context, you're not burning context tokens to do the extraction yourself.

Question 5

How is this different from Foundation?

Accepted Answer

Different jobs. Unstructured Transform parses files you point at right now; Foundation gives your agents standing access to your whole corpus across sessions.

Question 6

Do I lose control over how files get parsed?

Accepted Answer

No. Your agent chooses the best strategy based on the tool's instructions, but you can override it in plain language: force hi-res or VLM parsing, table enrichments, chunking, embedding, and output formats.

Question 7

Will it handle my weird files? Scanned stuff, handwriting, ugly tables.

Accepted Answer

That's what it's for: 60+ file types including scans, images, spreadsheets, slide decks, and PDFs with charts or margin handwriting. Tables come back structured and layout survives.

Transform MCP is now available for Claude

Table Accuracy

Content Accuracy

Lowest Hallucination Rate

Point. Parse. Done.

Works with

Built for the
hard files.

It just works!

You point.

Transform decides.

Files are done.

Use Cases

Structured Data Extraction

RAG

Agents

Automation

Start today.

FAQs

Transform MCP is now available for ClaudeClineCodexCursorDevinIBM Bob