Ocrbase.dev – pdf→.md/.json OCR for developers
94.5% accuracy, self-hostable, open source—beats Textract on cost and accuracy.

Cheapest OCR API at $1 per 1,000 pages, but it's just a hosted wrapper around PaddleOCR.
Backend developers building document processing pipelines
Amazon Textract · Google Document AI · Nanonets
If we did the pricing math right (which I hope we did), it should be the cheapest OCR there is, at just $0.001/page. Let us know if you find it useful
94.5% accuracy, self-hostable, open source—beats Textract on cost and accuracy.
Replaces O(n²) token re-parsing with true O(n) streaming; Vercel SDK does 4K re-parses on 50KB payloads.
Fixes WHOIS's broken successor: one endpoint, consistent schema, handles multi-server routing automatically.
ActivityStreams 2.0 normalization means build once, add platforms without code changes.
Beats manual SQL macros at flattening nested JSON for dbt pipelines.
Yet another OCR API wrapper when JinaAI and Firecrawl already exist.