I Built LocalLedger: An Offline Receipt Organizer That Runs Gemma on Your Own Laptop

Описание и применение

LocalLedger – автономный инструмент для организации чеков и выставления бухгалтерских отчётов, работающий полностью локально без подключения к интернету, защищая данные клиентов и устраняя ежемесячные подписки.

Задача

Freelancers need an affordable, privacy‑preserving way to organize receipts and generate accounting reports without relying on cloud OCR services that charge $15–40/month and expose sensitive client data.

Результат

LocalLedger delivers a zero‑cost, fully offline receipt‑organizing solution that outputs accounting‑ready CSVs and a detailed quarterly review, protecting client data and eliminating subscription fees.

Рабочий процесс

• Point LocalLedger at a folder of receipt files.
• Run the CLI command python -m localledger scan <folder> to parse receipts using the deterministic‑offline engine.
• Optionally run python -m localledger scan <folder> --no-llm to skip the LLM narrative.
• Generate a CSV and HTML report in the out/ directory.
• Review the quarterly‑review.html for totals, category breakdowns, and accountant questions.
• Import ledger.csv into preferred accounting software.
• Use the local web console via python -m localledger serve for a printable review page.
• Fine‑tune or swap the LLM model via Ollama or an OpenAI‑compatible server on localhost.
• Run tests with python eval.py to ensure parsing, categorisation, and fallback behaviour.

Доказательства

LocalLedger is an offline invoice and receipt organizer for freelancers. You point it at a folder of receipts; it reads them on your own laptop, sorts every line into a category, writes an accounting‑ready CSV plus a plain‑language quarterly review, and ends with a list of questions to ask your accountant.

Источник

The entire thing runs locally on an open‑weight model, so the numbers never move and the cost per receipt is zero.

Источник

Zero third‑party dependencies — the core, the CLI, the web console and the test suite all run on the Python standard library alone.

Источник