Dharma Field Extractor
What it is
A 2-billion parameter model designed to extract specific, structured fields from document images into JSON format. It supports both open-ended extraction of all detectable fields and schema-directed extraction where the user defines the exact output structure.
The problem it solves
General-purpose LLMs are expensive and slow for high-volume, well-defined document field extraction, while standard OCR provides unstructured text that requires additional coding to map to specific fields.
What you could use it for
- Extracting line items and totals from supplier invoices
- Parsing patient details from medical requisitions for scheduling
- Mapping fields from ID cards to employee records
- Extracting parties and terms from contracts for compliance systems
Why it matters
Provides a cost-effective and faster alternative to frontier LLMs for high-volume document structuring tasks, enabling direct integration into downstream databases without manual mapping code.
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.