Reducto Releases r-1: A Single Pass Document Parsing Model That Cuts Errors 20% at 1 Cent Per Page
1 cent per page. Reducto's r-1 cuts document parsing errors 20% and costs 66-83% less than the multi-stage pipeline it replaces.
Why it matters
A single-pass document parsing model that consolidates OCR, layout, tables, and formatting into one inference step, with meaningful cost and accuracy gains. Practitioners using document AI should evaluate whether the 20% error reduction and 3-6x cost cut justify switching from their current pipeline.
The key facts
13 to knowr-1 released September 1, 2026
20% error reduction reported
Pricing: 1 cent per page (vs. legacy 3-6 cents per page)
Consolidates OCR, layout detection, tables, formatting, and grounding into single pass
Replaces multi-stage agentic pipeline
Cost reduction: 66-83% cheaper than prior approach
Release date: September 1, 2026
20% error reduction vs. prior version
Pricing: $0.01 per page (flat rate)
Legacy pricing: $0.03–$0.06 per page
Cost savings: 66–75% reduction
Single-pass architecture: replaces multi-stage agentic pipeline
Unified capability: OCR, layout detection, tables, formatting, grounding in one model
Go to the source
MarkTechPostmarktechpost.com
Publisher excerpt: We look at r-1, the document parsing model Reducto released on September 1, 2026. We walk through how it folds OCR, layout detection, tables, formatting and grounding into one full page pass, replacing the multi stage agentic pipeline it ships alongside. We break down the two numbers that matter…