Readable instruction packs
Approve the extraction logic before the first live run.
Source-linked document extraction
Extract data from PDFs. Verify it at the source.
Extract fields from multiple PDFs and review available source evidence. Download a structured CSV, ready for analysis and further processing.
One trial per account. Card required. Includes 20 pages; downloads unlock on paid plans.
This silent demonstration shows multiple PDFs being uploaded, extraction fields being defined and approved, results being checked against highlighted source evidence when available, and structured data being downloaded as a CSV for further processing.
Flat or relational output
Handle one-row summaries or repeated records in the same workflow.
Source-linked review
Review available source pages, quotes, and exact or approximate highlights behind extracted answers.
Workflow
From upload to source review in one continuous flow.
Keep setup, approval, and evidence review inside one product instead of spreading the workflow across separate tools.
01
Upload the document set
Start with one PDF or a clean batch for the same job.
02
Define the fields that matter
Set columns, hints, and the output mode in plain language.
03
Approve the instruction pack
Review and edit the generated rules before the run starts.
04
Review evidence, not just rows
Jump from a result cell to the supporting PDF evidence.
Document types
Built for the PDFs teams actually have to work through.
Support structured numeric files, research-heavy reports, and mixed operational packets without changing the review model.
Invoices, contracts, and estimate-style PDFs
Pull totals, dates, clauses, and repeated line items without losing context.
Invoice numberDue dateLine itemsTotal amount
Learn more about PDF table extraction →Clinical studies, journals, and narrative reports
Extract findings, cohorts, endpoints, and narrative evidence from dense prose.
Study designSample sizeEndpointAdverse event
Learn more about research paper extraction →Compliance binders, forms, and mixed document sets
Pull IDs, dates, statuses, and policy details from mixed document packets.
Policy IDEffective dateStatusReference clause
Learn more about public PDF extraction →One 14-day free trial
Test up to 20 pages on the paid plan you choose.
Public interactive sample
Explore a complete sample extraction without signing in or entering a card.
Source-linked evidence
Grounded results can include the original PDF page, quote, and highlight.
Auto-deleted after retention
Uploaded documents are scheduled for automatic deletion after the retention window.
Trial & Pricing
Choose a paid plan for your one 14-day free trial.
The trial is available once per account. A card is required, and downloads unlock after paid billing begins.
Choose a paid plan for your one 14-day free trial.
Starter
$19/mo USD
$19 every month
For light monthly extraction work and occasional document batches.
- 100 pages each month
- One 14-day trial per account
- 20 trial pages total; card required
- CSV export unlocks when paid billing begins
- Billing starts after the trial unless canceled
- Quarterly plans are billed every 3 months
One trial per account. Card required. Trial includes 20 pages; downloads unlock on paid plans.
Premium
$90/mo USD
$90 every month
For recurring operations that need higher monthly throughput.
- 1,000 pages each month
- One 14-day trial per account
- 20 trial pages total; card required
- CSV export unlocks when paid billing begins
- Billing starts after the trial unless canceled
- Quarterly plans are billed every 3 months
One trial per account. Card required. Trial includes 20 pages; downloads unlock on paid plans.
Professional
$600/mo USD
$600 every month
For teams running high-volume extraction on a steady cadence.
- 10,000 pages each month
- One 14-day trial per account
- 20 trial pages total; card required
- CSV export unlocks when paid billing begins
- Billing starts after the trial unless canceled
- Quarterly plans are billed every 3 months
One trial per account. Card required. Trial includes 20 pages; downloads unlock on paid plans.
FAQ
Common questions before you start.
Which workflow should I use?
Use Extraction for named fields from one or many PDFs. Choose Flat mode for one row per file, or Relational mode when each PDF can contain repeated records.
How should I set up my first extraction?
Start small with 2-3 fields you can verify quickly, using representative PDFs rather than the biggest batch.
How do I verify a result?
Select a populated or otherwise reviewable value to inspect any grounded source quote and page. Highlights may be exact, approximate, or unavailable, especially for scanned PDFs.
What files are allowed?
Upload PDF files only, up to 10 per run and 15 MB per file. Direct PDF works with normal text PDFs and scanned documents.
Is my data secure?
Source files require an authenticated account and job-ownership checks. Direct PDF sends the complete original document to the AI provider configured for extraction. Use public or non-confidential documents only; this is not a HIPAA-compliant environment or a GDPR compliance tool.
Get started
Start extracting structured data from your PDFs today.
Choose a paid plan and get 14 days free to test live extraction with your own documents. The trial is available once per account; a card is required.