# CosmosShadow/gptpdf

Using GPT to parse PDF

Repository: https://github.com/CosmosShadow/gptpdf
Canonical: https://ross.abutalabs.com/products/gptpdf
Language: Python
License: MIT
License Family: permissive
Last push: 2025-04-17T07:20:29+00:00

## Health v2 (maintenance only)
Score: 31/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 17, release rhythm 35, longevity 56
- inputs: {"age_days": 796, "days_push": 503, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3564, forks 260 (observed 2026-08-28T04:08:09.896090+00:00)

## What it is
A small Python library that parses PDF files into Markdown using a vision-capable LLM such as GPT-4o. It uses PyMuPDF to detect non-text areas and sends them to a multimodal model to handle formulas, tables, charts, and images.

## Use cases
- convert pdf to markdown
- extract math formulas from pdf
- parse tables and charts from pdf documents
- ocr pdf with gpt-4o
- convert research papers to markdown
- extract images from pdf while converting text

## When to choose
- you need high-fidelity PDF-to-Markdown conversion including formulas, tables, and charts
- you have access to an OpenAI-compatible vision model API and per-page cost (~$0.013) is acceptable
- you want a simple, minimal Python API (parse_pdf) rather than a heavy pipeline

## When to avoid
- you need fully offline or free parsing with no LLM API costs
- you must process large volumes of PDFs where per-page API charges are prohibitive
- you need pixel-perfect layout reproduction rather than Markdown output

## Facets
- artifact type: library
- maturity: active
- function: pdf, ocr, llm-inference, parser
- domain: pdf, large-language-models, developer-tools
- platform: python, cross-platform
- tags: pdf-to-markdown, vision-language-model, document-parsing, openai-api

## Member repositories
- CosmosShadow/gptpdf (main) score 31

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:09.896090+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:34:10.905734+00:00, confidence not recorded.
  - readme: https://github.com/CosmosShadow/gptpdf (fetched 2026-08-28T04:08:09.896090+00:00, sha 6fe2e5e57a76)
  - registry_pypi: https://pypi.org/pypi/gptpdf/json (fetched 2026-08-29T09:27:46.847729+00:00, sha 272ed6a1d07b)
- Data as of 2026-08-30T08:39:29.467469+00:00.