# Changelog of Documents to Markdown for LLMs: PDF, Office, HTML, OCR (`rado-dev/documents-to-llm-markdown`) Actor

- **URL**: https://apify.com/rado-dev/documents-to-llm-markdown/changelog.md
- **Full Actor documentation**: https://apify.com/rado-dev/documents-to-llm-markdown.md

## Changelog

### 0.1 (unreleased)

- First version: PDF (text layer + tables), DOCX, PPTX, XLSX, HTML, TXT/MD, CSV, JSON, XML, images.
- Tesseract OCR (auto / always / never) for scanned PDF pages and images; 10 languages bundled.
- Token counts (cl100k_base) and heading/page-aware chunking for RAG.
- Pay-per-event billing: document, page, ocr-page. Failed and empty documents are not billed.
- Respects the run's maximum cost: pages that cannot be paid for are not converted.
- Office/HTML conversion without MarkItDown (same libraries underneath) - ~2 s less start-up CPU per run, smaller image.
- Key-value store input via resource picker (works under limited permissions).