Choosing fully-local batch PDF OCR in 2026: docling, GLM-OCR, PaddleOCR-VL, olmOCR 2, MinerU
TangeloOk9486 · reddit · 2026-11-02
The author needs a fully local pipeline for batch PDF processing (20–25 pages each, mostly text, some tables and photo pages) and shortlists docling (granite-docling), glm-ocr, PaddleOCR-VL, olmOCR 2, and MinerU.
Context and trade-offs: company policy requires files to stay on their own servers, ruling out LlamaParse's enterprise self-host route and cloud services; LiteParse from the same team runs locally and is fast on plain text, but its rule-based Markdown with tesseract fallback struggles with merged cells and multi-column pages. Two open practical questions: is a VLM overkill for text-heavy PDFs (vs. routing only table/image pages to a VLM), and should VLMs get the whole 25-page PDF at once or page batches?
More from Infra
- CVSS 9.8 ingress-nginx flaw lets unauthenticated attackers get RCE and steal cluster Secrets — evilsocket · 2026-11-02
- RAM prices tripled, so Tesla cut a third of memory from Optimus's brain — aakashgupta · 2026-11-02
- Liquid AI's 1.7B LFM3 becomes first local model to nail live fixtures lookup on a MacBook — NielsRogge · 2026-11-02
- To dethrone Postgres, new databases must prioritize ecosystem integrations from day one — mgill25 · 2026-11-02
- ESP32-powered 4-inch touchscreen builds a TUI dashboard for monitoring a DGX inference cluster — CATLLM · 2026-11-02
- OpenRouter crosses 10 quadrillion annualized token run rate, doubling every 10.5 weeks — deedydas · 2026-11-02