Reducto has launched r-1, a single-pass document parsing model designed to consolidate text, tables, figures, and layout analysis into one process. The model is currently available in preview via Reducto's hosted Parse API on V3, accessible through a configuration flag. There are no open weights or local checkpoints for self-hosting.
The company reports a 20% reduction in error rates compared to its own legacy agentic pipelines. In internal evaluations, r-1 outperformed Amazon Textract and Azure Document Intelligence on complex documents. The model natively handles digital text, scans, handwriting, and complex table structures, including merged cells and nested headers.
Pricing has been significantly reduced to $1 cent per page, down from the $3-6 cents per page charged for legacy agentic models. This shift is part of a wider move toward flat product rate cards. Reducto also offers up to $5,000 in credits for organizations migrating from other parsers to conduct side-by-side comparisons.
Future updates include the release of r-1 mini for speed and cost-sensitive workloads, as well as automatic per-page routing to select the optimal model for each page. For workflows requiring custom prompts or advanced chart extraction, Reducto continues to support agentic processing as an augmentation to r-1 results, though this adds latency.
