Noesion Radarnoesion.ai
Release2026-10-02

Datalab released OmniExtractBench. It is a structured extraction benchmark. It uses 620 diverse documents from four suites. It provides clear, consistent and auditable scoring. It tests edge cases like scans, dense tables and forms. The dataset is CC BY 4.0 licensed. The scorer is Apache 2.0 licensed. It can run with one command line. It supports multiple providers. It returns accuracy, precision, recall and F1 metrics. It offers verdicts for deeper failure analysis.

What it unlocks

Benchmarking structured extraction models on realistic, auditable, diverse documents.

  • 620 documents
  • 4 suites
  • CC BY 4.0 license

What you need to act on it

  • Python environment
  • API keys for providers

Sources

Send this to someone who needs it

Shares the story and its sources. Nothing about you.

What does this mean for your job?

This is the story as everyone gets it. Once a week we send you the version written for your role — what changed, why it matters for the work you actually do, and one thing to try. Free while we tune it.