No Chinese models allowed: what's left for local document OCR when contracts rule out PaddleOCR and Qwen?
aidenclarke_12 · reddit · 2026-11-08
A Reddit user needs fully local processing of scanned inspection reports and material certs — splitting stacks and extracting lot numbers/alloys into a database — but their contract forbids Chinese software and models, eliminating PaddleOCR and Qwen (even olmOCR 2 is a Qwen fine-tune). Remaining candidates: granite-docling, liteparse, tesseract and gemma. They ask for recommendations and how to demonstrate model provenance to customers.
More from coding & agent
- DataTalksClub frees course materials for Production AI Agents for Generative Media, projects application-only — Al_Grigor · 2026-11-08
- Engineer's data: 11x more merged PRs per month at AI-native Tano vs pre-AI PostHog — bibryam · 2026-11-08
- ADLC means two different things: Agentic Development Lifecycle vs Agent Development Lifecycle — Thirumalaivasan_GJ · 2026-11-08
- Anthropic's Official claude-quickstarts Repo Nears 20k Stars for Claude API Developers — anthropics · 2026-11-08
- Open-source rwxmap labels API operations r/w/x with a 23MB local model replacing a paid LLM — Tight_Heron1730 · 2026-11-08
- Working With AI: Getting Better Results by Being 'Dumber', Not Smarter — glcst · 2026-11-08