Consoleby Nace.AIIntroducing Nace Document Intelligence: The Perception Layer for Agents
October 8, 2026
Introducing Nace Document Intelligence: The Perception Layer for Agents
Today we’re releasing Document Intelligence in the Nace Console, bringing frontier document processing directly into the Nace stack.
Nace Document Intelligence (NDI) is our Perception Layer, while Drex is our Decision Layer. Together, they give agents a suite of tools to understand files and make decisions on top of the information inside them.
NDI can parse, split, classify, extract, and ground 36 file types, including PDFs, scans, spreadsheets, emails, images, audio. Every result stays connected to its source, whether that means a page and bounding box, spreadsheet cells, a text span, or a timestamp.
Agents are only as good as what they can read
Most real workflows don’t start with clean text. They start with invoices, contracts, financial statements, spreadsheets, scans, emails, and mixed document packets.
Traditional OCR can recover text but often loses the structure around it. General-purpose frontier models can understand complex documents, but using them to process every page quickly becomes expensive.
NDI is built specifically for this layer, turning unstructured files into structured, grounded information that downstream agents can reliably use.
Five document processing tasks, one API
The Perception Layer gives builders five tools through API, CLI, SDK and MCP interfaces.
Parse converts files into structured, layout-aware output while preserving tables and source locations. Split separates mixed packets into separate documents inside them. Classify labels documents or pages using any provided taxonomy with confidence levels. Extract returns structured fields against a schema, with exact location and confidence levels. Ground resolves information back to the exact place it came from.
Frontier performance at a fraction of the cost
On our parsing evaluation, NDI Parse matched the best overall score we tested at 0.820, while costing less than half of LlamaParse Agentic and roughly 10× less than GPT-6 Astra.
Model | Overall | OmniDocBench | ParseBench | Cost / 1K pages |
Nace DI High | 0.820 | 92.6 | 91.16 | $6.50 |
LlamaParse Agentic | 0.820 | 89.4 | 91.78 | $13.80 |
GPT-6 Astra | 0.813 | 89.6 | 90.67 | $70.57 |
NDI Parse | 0.805 | 93.8 | 89.05 | $3.50-7.00 |
Mistral OCR 4 | 0.796 | 89.2 | 89.55 | $4.00 |
Anthropic Opus 4.7 | 0.787 | 88.2 | 90.26 | $71.40 |
For document-heavy agents, this trade-off matters because document processing happens before every downstream reasoning or decision step. At scale, accuracy, latency, and cost compound quickly.
Perception + Decision
NDI reads and structures the information inside your files. Drex takes that information and decides what to do next.
Both live in the same Console, use the same API key, and can be combined in a single workflow, giving builders a direct path from raw files to grounded decisions without stitching together separate systems.
Document Intelligence is available now in the Nace Console.