Tag
Datalab has released Marker v2, an open-source document parsing pipeline that efficiently converts PDFs, images, DOCX, and PPTX files to markdown with support for over 90 languages and high performance on GPUs.
The author tested DokieAI's presentation plugin for Codex and Claude Code, which converts Markdown into team-ready presentations via MCP, sharing before/after results.
A Twitter thread promotes crawl4ai, an open-source web crawling tool for LLMs that converts any URL into LLM-ready markdown, offering free unlimited access compared to paid services like Firecrawl, ScrapingBee, and Apify.
Ben Overmyer created a Python script using Cursor to recover and convert legacy Windows Live Writer (.wpost) files into Markdown format, including embedded images. The tool has been open-sourced on Codeberg for others to use.