Developments

Company announcements and posts from the past 30 days.

18 updates · By announcement or publication date

Project	Update	Date
datalab-to/indra	API submits and result checks, 2x faster ↗Company post · datalab.to	Nov 6, 2026Published
datalab-to/indra	Which document-parsing benchmarks can you trust? ↗Company post · datalab.to	Nov 4, 2026Published
datalab-to/indra	Data extraction vs parsing: Which to use? ↗Company post · datalab.to	Nov 2, 2026Published
datalab-to/indra	Surya 2.5: 1.7x faster, best open OCR under 1B ↗Company post · datalab.to	Oct 29, 2026Published
datalab-to/indra	Chandra 2.6: more accurate than the frontier, 2.5x faster, same price ↗Company post · datalab.to	Oct 27, 2026Published
datalab-to/indra	Lite vs Azure, AWS and Google: 2x faster, 10x cheaper, as accurate ↗Company post · datalab.to	Oct 26, 2026Published
datalab-to/indra	Open-source OCR models on OmniParseBench ↗Company post · datalab.to	Oct 23, 2026Published
datalab-to/indra	Lite mode: document parsing for $0.50 per 1,000 pages ↗Company post · datalab.to	Oct 22, 2026Published
datalab-to/indra	Improved document segmentation ↗Company post · datalab.to	Oct 21, 2026Published
datalab-to/indra	Word boxes that hold up on handwriting ↗Company post · datalab.to	Oct 20, 2026Published
datalab-to/indra	Find the fillable fields on any form ↗Company post · datalab.to	Oct 19, 2026Published
datalab-to/indra	Indra OCR 1.1: handwriting, 80+ languages and pages at any angle ↗Product · datalab.to	Oct 19, 2026Published
datalab-to/indra	Script confusions: when an OCR model writes the right word in the wrong alphabet ↗Company post · datalab.to	Oct 16, 2026Published
datalab-to/indra	Managing Model Uncertainty to Catch OCR Hallucinations ↗Company post · datalab.to	Oct 15, 2026Published
datalab-to/indra	Chandra 2.5: Structured Image Descriptions and Faster OCR ↗Company post · datalab.to	Oct 14, 2026Published
datalab-to/indra	How KIQP's technical team processed 30,000 pages of historical Arabic archives ~65% faster ↗Company post · datalab.to	Oct 13, 2026Published
datalab-to/indra	Indra OCR: a 1M-parameter OCR model that reads 49 pages per second on CPU ↗Company post · datalab.to	Oct 13, 2026Published
datalab-to/indra	Why we built OmniParseBench ↗Company post · datalab.to	Oct 12, 2026Published

About this feed

Dates use the announcement date when known, otherwise the source publication date. Company posts come from official sources. Other reporting is reviewed for a company match. Coverage depends on accessible public sources.