LandingAI Releases Agentic Document Extraction Gen2 With DPT-3 Pro and DPT-3 Verity

The release splits parsing into two models. DPT-3 Verity, in public preview, transcribes digitally created documents deterministically and returns a bounding box and confidence score for every word; it targets high-volume text, tables and simple form fields. DPT-3 Pro reads page layout before words, detects block types from tables and figures through marginalia and signatures, returns them in reading order, and handles scanned pages, handwriting, non-Latin scripts and LaTeX math. LandingAI said Verity charges roughly 40% of the credits Pro charges and that automated routing between the two is planned for fall 2026.

The pricing change is central. Under DPT-2, every page cost a flat 3 credits. Under DPT-3, credit consumption is the sum of a page component and an output character component. On the priority tier, DPT-3 Pro bills 1 credit per page plus 0.5 credits per 1,000 output characters. DPT-3 Verity bills 0.3 credits per page plus 0.2 credits per 1,000 output characters. The standard tier halves both rates, so a 12-page Pro parse returning 48,120 characters comes to 36.1 credits on priority and roughly half that on standard. Totals round up to the nearest 0.1 credit, and response metadata reports every input to the calculation.

Service tiers are the second pricing lever. Priority is for when a person or an agent is waiting. Standard runs asynchronously at 0.5x price and suits pipelines that tolerate minutes to hours. Synchronous calls always bill at priority, and the playground itself runs on priority. LandingAI projects 25% to 80% cost reductions on mixed workloads and claims parsing under one cent per page with Verity on standard. Those are vendor figures until customers benchmark their own document mix, because the character component means a dense page can cost more than it did before.

The Parse v2 response has three top-level fields: markdown in reading order, metadata, and structure. The structure is a document node whose children are pages, whose children are blocks. Block types include text, table, table_cell, figure, marginalia, attestation, logo, card and scan_code. Every block carries a semantic ID in the form type-index, stable within a response but not across re-parses, plus a grounding object with the page number, a range into the markdown string, and a normalized bounding box.

Markdown output is standardized. Figures use style elements with generated prose isolated inside tags, so transcription is never mistaken for model commentary. Attestations emit stacked labels such as [STAMPED][SIGNED], with [ILLEGIBLE_SIGNATURE] and [ILLEGIBLE_TEXT] as fixed literals. Tables ship as HTML by default to preserve merged cells.

Atomic grounding is the release's most consequential capability, according to the report. Every leaf block carries an atomic_grounding array: one entry per visual line with DPT-3 Pro, one entry per word with DPT-3 Verity. Verity attaches a confidence value from 0 to 1 per word, computed as the lowest per-character score in that word, giving teams a signal for routing uncertain transcriptions to review. Table cells now carry their own bounding boxes, though Pro leaves cell-level atomic grounding empty. Extract V2 draws citations from this grounding, so an extracted field traces to a specific word on a specific page. That makes PII redaction by coordinate, document diffing and reviewer UIs buildable rather than approximate.

Gen1 client code will not run against Gen2 endpoints, so migration is required, the report said. Developers can start free in the ADE playground.

Editor's Summary

LandingAI has made ADE Gen2 generally available, replacing DPT-2 with DPT-3 Pro and DPT-3 Verity and switching parsing to page-plus-output-character pricing. The release adds block-level document structure and atomic grounding down to lines or words, with Verity adding per-word confidence scores. Gen1 clients must migrate because the new endpoints are not compatible.