Documentation
Start extracting document data with AI
Upload, review, and export your first file
Dynamite Docs- Upload your first file
- Sign in with your account for higher limits and cloud storage (on paid plans); anonymous visitors can try extraction with tighter, per-IP daily caps.
- Run extraction
- Drag and drop a document anywhere into the file panel, or press Ctrl+U. The file appears in the sidebar immediately.
Dynamite Docs- 01Upload your first file
- 02Run extraction
- 03Review the result
Dynamite Docs turns business documents such as invoices, receipts, contracts, statements, and forms into structured, editable tables. There is no template setup and no training: you upload a document, the AI infers the fields, and you correct anything it missed. It saves those corrections as patterns that can be reused for a similar layout.
Prerequisites
- A supported document file (PDF, PNG, JPG, TIFF, WEBP, DOCX, XLSX, or EML) under 50 MB.
- A desktop or mobile browser with JavaScript enabled (Chrome, Edge, Firefox, Safari).
- Internet connection for hosted AI models (or active local Ollama companion for local inference).
- Free account sign-in to save layout patterns and download CSV, JSON, or XLSX exports.
Sign in to your account
Click the [Sign In] button in the top navigation bar to authenticate with Clerk, or start immediately as an anonymous guest. Signed-in users receive higher processing limits and persistent cloud storage on paid tiers.
Upload your document
Click the [Upload] button in the Library sidebar or press Ctrl+U (⌘+U on macOS). You can also drag and drop your document file directly into the left sidebar file list.
Run schema inference and extraction
Select the uploaded document from the Library sidebar, then click the [Extract] button in the Data Studio toolbar or press Ctrl+Enter. The AI reads the page and populates the structured table below.
Review and correct values inline
Verify extracted numbers and text against the source document. Double-click any cell to edit mistaken text or amounts directly. Click the [Save] button or press Ctrl+S to commit changes and store the layout pattern.
Export verified structured data
Click the [Export] button in the Data Studio toolbar or press Ctrl+Shift+S, then select [Download CSV], [Download XLSX], or [Download JSON].
Exact error messages & resolutions
Document exceeds plan limit of 5 pagesSplit multi-page PDFs to 5 pages or fewer, or upgrade to Starter (up to 50 pages) or Pro (up to 500 pages).
413 Payload Too Large (50 MB limit)The uploaded file exceeds the 50 MB server boundary. Downsample high-DPI image scans or compress the PDF.
Monthly processing limit reached (0 PE remaining)All page-equivalents for the current billing cycle have been spent. Processing resumes at monthly reset or upon upgrading your plan.
Troubleshooting this task
Table returned empty or missing lines: Check if your PDF is an encrypted or password-protected file. For physical paper scans, confirm the document is oriented upright and scanned at 150 DPI or higher.
Wrong columns inferred: Double-click column headers in Data Studio to rename fields or click [Add column]. Once saved with [Save], the layout memory will remember your structure for future files.
Related workflows & guides
- Learn about supported raster, vector, and office file formats inUploading documentation.
- Understand how the infer, correct, and remember cycle operates in theExtracting data guide.
- Compare hosted page allowances and cloud storage options acrosssubscription packages and limits.
Dynamite Docs- Upload your first file
- Sign in with your account for higher limits and cloud storage (on paid plans); anonymous visitors can try extraction with tighter, per-IP daily caps.
- Run extraction
- Drag and drop a document anywhere into the file panel, or press Ctrl+U. The file appears in the sidebar immediately.
- Review the result
- Select the file and press Ctrl+Enter (or click Extract in the Data Studio) and the extracted table lands in the panel below.