LandingAI Releases Agentic Document Extraction Gen2

LandingAI has officially released Agentic Document Extraction Gen2 (ADE Gen2), which is generally available now. Developers can start for free in the ADE playground, while enterprises can deploy it across US or EU clouds, within private VPCs on AWS, Azure, or Google Cloud, inside Snowflake, or on-premises in air-gapped environments.

Gen2 splits document parsing into two distinct models to match workloads with appropriate pricing. DPT-3 Verity deterministically transcribes digitally created documents, returning a bounding box and confidence score for every word, targeting high-volume text, tables, and simple form fields. DPT-3 Pro reads page layouts before processing words, detects block types including tables, figures, marginalia, and signatures, returns them in reading order, and handles scanned pages, handwriting, non-Latin scripts, and LaTeX math. LandingAI notes that DPT-3 Verity charges roughly 40% of the credits required by DPT-3 Pro, with automated routing planned for fall 2026.

Stock photo for illustration only, not from the actual event

Under the previous DPT-2 model, every page incurred a flat 3-credit fee. Under DPT-3, credit consumption is calculated as 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, whereas DPT-3 Verity charges 0.3 credits per page plus 0.2 credits per 1,000 output characters. The standard tier halves both of these rates.

Service tiers serve as an additional cost control lever, with the priority tier intended for real-time human or agent workflows, and the standard tier running asynchronously at half the price for pipelines tolerating longer processing times. LandingAI projects cost reductions between 25% and 80% on mixed workloads.

Source: MarkTechPost