MinerU turns PDFs and Office docs into LLM-ready Markdown to cut token costs
lxfater · x · 2026-07-28
MinerU is a 75.9k-star open-source project that converts PDFs, images, and Office documents into LLM-ready Markdown/JSON for agent workflows.
The post argues that uploading PDFs directly to an AI can burn roughly 4× more tokens because the model has to use visual processing to preserve layout and tables. MinerU aims to cut that cost by extracting document structure while keeping key visual and formatting information, with a web version that removes deployment friction.
More from coding & agent
- Snapshield snapshots a git repo before an AI coding agent session and restores it with one command — Ok-Conversation238 · 2026-07-28
- Hydra routes local tasks to the cheapest model that clears a confidence threshold — jhaankit373 · 2026-07-28
- Startup playbook says to ship an MVP with Claude Code, charge on day one, and automate marketing — sahilypatel · 2026-07-28
- Builder’s job is orchestration, not handing judgment to the model — cornmacabre · 2026-07-28
- Open-source code review agents are now running inside GitHub Actions — Silly_Entertainer92 · 2026-07-28
- Do panel-of-judges code reviewers still matter as agent models improve? — NeighborhoodOwn8510 · 2026-07-28