01
Infer the structure
The model separates document fields, repeating rows, totals, and notes based on the file in front of it.
What is included in the image JSON export?
Dynamite Docs uses AI OCR to extract labelled fields and table rows from a document photo, scan, or screenshot. Review the values beside the image, then export JSON for your mapper.
What to expect
Valid JSON can still contain an OCR mistake. Keep identifiers as strings and validate required fields, dates, amounts, arrays, and null handling in your application.
The full workflow
The output stays beside the evidence.
Check the proposed fields against the source, correct a value, then export the reviewed structured image JSON.
Dynamite Docs- 01Select the source file
- 02Review fields and rows
- 03Download structured image JSON
02
Correct against the source
The source remains visible while you edit cells, rename columns, inspect warnings, and mark the result reviewed.
03
Send useful data onward with Dynamite Docs
Download structured image JSON, or move a recurring workflow into Google Sheets, Drive, webhooks, or the API.
Dynamite Docs- document_type
- service_record
- fields
- Named values outside the table
- columns
- Definitions for repeating row values
Choose the AI route
Bring a free provider key. Keep the review workflow.
Open AI & Processing, choose Connect Provider, and pick a provider. The setup screen links to that provider's key page. Create a free key where the provider offers one, paste it into Dynamite Docs, verify it, then select a model for the run.
Your provider key is encrypted before it is stored.
Switch between hosted, own-key, privacy-restricted, and local models.
The same source-first review screen works whichever route you choose.
Choose the page for the job using Dynamite Docs
Use Image to JSON when developers need structured JSON fields and row arrays from a document photo, scan, or screenshot. Capture high-resolution images with uncropped margins so field labels remain clearly associated with their corresponding values.
During review, cross-check extracted key-value pairs and table rows against the source image. Verify that alphanumeric identifiers retain leading zeros as strings, that dates follow standard formats, that decimal amounts avoid floating-point distortion, and that missing values are mapped to consistent null representations.
Use Image to CSV for flat tabular imports, or Image to Excel for spreadsheet analysis. When the image is a commercial invoice, use Invoice to JSON for specialized vendor, tax, and line-item mapping. For recurring programmatic image ingestion, connect via our REST API with saved schemas, webhook callbacks, and retry handling.
The 28-day Free trial includes 25 page-equivalents per month shared by hosted and own-key processing, 5 hosted pages per document, and 5 documents per day. Anonymous trials have separate daily limits.
Setup, review and export questions
Does Image to JSON encode the picture as Base64?
No. This tool extracts visible document data into fields and rows. It does not serialize the original image bytes into a JSON string.
Can I choose my own JSON property names?
Review and rename inferred columns in the workspace. For a repeatable production schema, save a preset and validate the exported values in your application.
When should I use the document extraction API?
Use the free page to inspect one file and a sample output. Use the API when software needs authentication, repeatable schemas, retries, and automated delivery.
How should I handle OCR confidence in an application?
Use it to route fields for review, not as proof that a value is correct. Define which identifiers, dates and amounts always need source comparison, and send missing or unexpected fields to an exception path instead of accepting the payload silently.
Why should identifiers remain strings?
Work-order numbers, account references, postal codes and similar values may contain leading zeroes, punctuation or letters. Numeric conversion can alter them without producing an error. Keep the reviewed representation as text and validate its business format separately.
Process the whole folder in one run.
Starter and above add stored files, batch jobs and connected exports. Email intake requires Pro or Scale. File and processing limits depend on your plan.
Compare batch plans