Choosing the Right Parser Type

Parsio currently supports 3 parser types:

Choosing the appropriate parser for parsing your documents is crucial.

The 3 tables below show the main differences, advantages, and limitations of each, making it easier for you to choose the right parser for your use case.

Parser	Use cases	Parser description	Cost
Template-based	Parsing transactional machine-generated emails with a fixed layout. Parsing structured file formats: XML, CSV, Excel, JSON.	Create a parsing template by highlighting the data to extract. Structured file formats are processed automatically.	1 credit / document
AI PDF parser	Parsing PDF files: invoices, receipts, business cards, ID documents, PDF tables, W-2 forms and general documents.	Pre-trained AI models automatically extract all available data.	5 credits / page
GPT-powered parser	Parsing complex emails, PDFs and documents where other parsers failed.	Write a text prompt as if you were talking to a person.	2 credits / page

Parser	Supported formats
Template-based	Emails, PDF, HTML, Excel, CSV, XML, DOCX, JSON
AI PDF parser	PDF, JPG, PNG, BMP, TIFF
GPT-powered parser	Emails, PDF, HTML, TXT, DOCX, XML, MD, JSON

Parser	Advantages	Limitations
Template-based	Easy to use. Perfect for transactional emails and simple text PDFs. Built-in templates: Haro, Airbnb, Etsy, LinkedIn and more.	No OCR for PDFs. Only some simple text PDFs. Cannot extract tables from PDF files.
AI PDF parser	OCR: parse text and scanned PDFs. Extract tables from PDFs. Parse hadwritten text. Uses pre-trained AI models: no need to create parsing templates or parsing rules.	Limited to avalable pre-trained models: invoices, receipts, business cards, ID documents, PDF tables, W-2 forms and general documents.
GPT-powered parser	Very powerful for parsing emails, PDFs and documents with complex layout where creting of a parsing template is impossible. Easy to use: write a text prompt as if you were talking to a person.	Sometimes it cannot extract large chunks of data. In such cases, it's better to split your document into multiple pieces.