Unstructured’s New Transform Models Beat GPT-6, Claude and Gemini at Parsing Tables, Forms and Scans

Unstructured, whose open source software recently surpassed 100 million downloads, launched new models for Transform, its document parsing solution for turning PDFs, scans and other complex documents into data AI agents can act on. In testing on 224 real-world business pages, GPT-6, Claude and Gemini produced up to 2.3 times as many table errors as Transform and invented about 50% more content than Unstructured that was never in the document.¹ The new models are live in Transform now with 10,000 free pages at transform.unstructured.io.

Also Read: AiThority Interview with Michael Kwok, Vice President, IBM Bob, and Canada Lab Director

Tables, forms and scans are where AI fails quietly. A decimal point drops out of a table cell, and $1,407.77 becomes $140,777. An empty checkbox on an insurance claim is read as checked. A blank field on a loan application comes back filled with a hallucinated guess. When a person reviews the output, someone catches the mistake. When an agent acts on it, the error flows straight into a payment, a decision or an answer to a customer.

The 224 pages come from SCORE-Bench, Unstructured’s open benchmark of real-world business documents, where Transform ranked first on every table-accuracy measure.¹ Transform costs $0.015 a page; GPT-6 costs about 7.5 times as much on the same run.

Transform also holds up on the benchmarks specialist parsers use. On RD-TableBench, the 1,000-table benchmark Reducto built to rank document parsers, Transform scored 0.899 using Reducto’s own grading script.² On forms, Transform now finds 89% of fields.³

“RAG needed mediocre data. Agents need extraordinary data,” said Brian Raymond, founder and CEO of Unstructured. “At 99% accuracy per step, a 50-step agent finishes clean 60% of the time. At 99.9%, it finishes clean 95% of the time. That extra point nine lives in the tables, forms and scans other tools get wrong. We built Transform to get them right. Every parser wins its own benchmark, ours included, so we’ll prove it on your documents.”

To back that up, Unstructured will run any team’s 25 hardest pages through the new models and return a free side-by-side comparison against their current parser within 24 hours, with a customized setup for any page that needs one. Teams moving to Transform from Reducto, LlamaParse, another parser or a frontier model can claim up to $10,000 in Transform credits through Dec. 31, 2026. Visit unstructured.io to learn more.

One call to Transform returns tables as real grids with merged headers intact, forms as label-and-value pairs with handwriting and checkbox states preserved, and scanned PDFs with no text layer as clean text, no separate OCR step needed. Every element comes back labeled by type, with its page number and coordinates, so an agent can cite exactly where an answer came from. In Transform’s workbench, users can get a recommended extraction schema from a sample document, simplifying schema iteration. Developers can choose between Balanced, the default parsing strategy, or Best for scans, handwriting and dense tables.

Transform is available self-serve at transform.unstructured.io or as a Dedicated Instance on AWS, Azure or GCP for teams that need infrastructure-level isolation or have specific data residency needs. Developers can start now at transform.unstructured.io with 10,000 free pages and no credit card.

Also Read: Why Coordination, Not Code, Is the Real Constraint on Shipping Software Today

[To share your insights with us, please write to psen@itechseries.com]