Explore / receipt-ocr

bhimrazy

receipt-ocr

Receipt-to-JSON in one pip install: CLI, Python API and FastAPI service that send a receipt image to any OpenAI-compatible model and return merchant, totals and line items, plus a Tesseract module.

636 79 Python MITupdated 28 days ago

ocr data

View on GitHubDispute this mapping →

Curator's take

Narrow on purpose, and that is the value — one schema, one document type, a working CLI, Docker service and FastAPI app on day one, and any OpenAI-compatible endpoint (OpenAI, Gemini, Groq) behind a single env var. Do not mistake it for a document-intelligence stack: no layout model, no bounding boxes, no batch pipeline, and accuracy is entirely whatever VLM you point it at. Good as a reference implementation or the expenses corner of a larger app; wrong choice for mixed document types or on-prem throughput.

Mapped by ShipWithAI editors · links verified

README.md

Receipt OCR Engine

An efficient OCR engine for receipt image processing.

This repository provides a comprehensive solution for Optical Character Recognition (OCR) on receipt images, featuring both a dedicated Tesseract OCR module and a general receipt processing package using LLMs.

Star History

Receipt OCR Engine

Quick Start

Extract structured data from a receipt in 3 steps:

Install the package:
```
pip install receipt-ocr
```

Set up your API key:

export OPENAI_API_KEY="your_openai_api_key_here"

Process a receipt:
```
receipt-ocr images/receipt.jpg
```

For Docker or advanced usage, see How to Use Receipt OCR below.

Project Structure

The project is organized into two main modules:

src/receipt_ocr/: A new package for abstracting general receipt processing logic, including CLI, programmatic API, and a production FastAPI web service for LLM-powered structured data extraction from receipts.
src/tesseract_ocr/: Contains the Tesseract OCR FastAPI application, CLI, utility functions, and Docker setup for performing raw OCR text extraction from images.

Prerequisites

Python 3.x
Docker & Docker-compose(for running as a service)
Tesseract OCR (for local Tesseract CLI usage) - Installation Guide

How to Use Receipt OCR

Receipt OCR Module (Structured Data Extraction)

This module provides a higher-level abstraction for processing receipts, leveraging LLMs for parsing and extraction.

To use the receipt-ocr CLI, first install it:

pip install receipt-ocr

Configure Environment Variables: Create a .env file in the project root or set environment variables directly. This module supports multiple LLM providers.

Supported Providers:
- OpenAI:
  
  Get API key from: https://platform.openai.com/api-keys
```
OPENAI_API_KEY="your_openai_api_key_here"
OPENAI_MODEL="gpt-4o"
```
- Gemini (Google):
  
  Get API key from: https://aistudio.google.com/app/apikey
```
OPENAI_API_KEY="your_gemini_api_key_here"
OPENAI_BASE_URL="https://generativelanguage.googleapis.com/v1beta/openai/"
OPE
```

Continue your stack

What teams reach for next — and why each earns a place beside receipt-ocr. Ranked by curator confidence.

Pairs well with

TurboOCR

GPU document parser in C++/TensorRT: PP-OCRv6 OCR, layout, tables and formulas to Markdown at 200-559 img/s on one GPU — no VLM, HTTP + gRPC, one-line Docker deploy.

Why it fitsIf per-image LLM calls stop scaling, TurboOCR does GPU-side text, layout and table extraction at hundreds of images per second, leaving only the schema-filling prompt for the model.

Alternative to See all + compare →

contextgem

Declarative LLM extraction from documents: describe Aspects and Concepts in plain language, get structured values back with paragraph- or sentence-level references and built-in justifications.

Why it fitsSame idea — describe the fields, get structured values back. ContextGem is a general declarative extraction framework with paragraph-level references and justifications; receipt-ocr hardcodes one domain so there is nothing to design before it works.

chandra

Datalab's SOTA open OCR model: images/PDFs to structured HTML/Markdown/JSON with layout, tables, forms, checkboxes, handwriting and math, in 90+ languages. Local HF or vLLM inference.

Why it fitsBoth read an image and return structure. Chandra is a purpose-trained OCR model handling arbitrary layout, tables and handwriting; receipt-ocr just prompts a hosted VLM against a fixed receipt schema, which is cheaper to adopt and brittler off-domain.

unstract

LLM-driven platform turning unstructured documents into structured data: a no-code Prompt Studio to define extractions, then deploy as APIs or ETL pipelines. Self-hosted, AGPL + enterprise.

Why it fitsBoth turn documents into structured data with LLMs and expose an API. Unstract is a no-code platform with Prompt Studio, ETL deployment and enterprise licensing; receipt-ocr is a 50-line integration you can read in full.