pdfmux
Self-healing PDF extraction with per-page confidence scoring. Open-source LlamaParse alternative for RAG pipelines, MCP server for Claude Desktop, LangChain + LlamaIndex loaders. Ranked #2 on opendataloader-bench (0.900).
The only PDF extractor that audits its own output. Catches blank pages, scrambled columns, broken tables — re-extracts them with a stronger backend. So your LLM gets clean data, not silent garbage. Routes each page to the best of 5 rule-based backends + BYOK LLM fallback (Gemini / Claude / GPT-4o / Ollama). One CLI. One API. Zero config.
PDF ──> pdfmux router ──> best extractor per page ──> audit ──> re-extract failures ──> Markdown / JSON / chunks
|
├─ PyMuPDF (digital text, 0.01s/page)
├─ OpenDataLoader (complex layouts, 0.05s/page)
├─ RapidOCR (scanned pages, CPU-only)
├─ Docling (tables, 97.9% TEDS)
├─ Surya (heavy OCR fallback)
├─ Marker (academic papers, neural)
├─ Mistral OCR ($0.002/page, 96.6% tables)
└─ YOUR LLM (Gemini / Gemma 4 / Claude / GPT-4o / Ollama / Mistral — BYOK via YAML)
