HTLinh0604/invoice_ocr_craft_llama3
This CRAFT + Llama 3.1 pipeline automates invoice semantic extraction, elevating accuracy from 65% to 94%. It achieves a 92% success rate in line-item reconstruction and is deployed via Flask API to streamline enterprise financial workflows.
Stars
1
Forks
—
Language
Jupyter Notebook
License
—
Category
Last pushed
Jan 13, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/HTLinh0604/invoice_ocr_craft_llama3"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
francesco-s/document-claim-mapping
A tool using LLMs and few-shot learning for document-claim mapping and evaluation. It extracts,...
sahajrajmalla/invoice-data-extraction-llm
Extract invoice document data using Large Language Models.
aryanesmaili/JobExtractor
a simple Job exteactor from job posting website that uses llama3.2 model to process and extract...
Anshu-312/llm_structured_extractor
Extract structured ticket fields from text using OpenRouter LLM with strict schema enforcement
veydantkatyal/doc-analysis
automatically extracts, summarizes, and analyzes PDF documents using Large Language Models...