Datalab Indra OCR: 1M params, 49 pages/sec on CPU
Why it matters
Indra OCR reads 49 pages/sec on 32 CPU cores with no GPU (11–17x vs Tesseract and PaddleOCR PP-OCR text models), outputs line boxes, layout labels and reading order as markdown/JSON, and beats PP-StructureV3 on OmniParseBench headline (62.1 vs 58.4) at ~1/180 the time per page. Code Apache 2.0; weights modified OpenRAIL-M.
Read the source: Datalab (opens in a new tab)
All hits from 14 Oct 2026More VisionGet the daily hits by email