multimodal benchmark
RealKIE-FCC
RealKIE-FCC is a key information extraction benchmark drawn from real enterprise documents (FCC filings), part of the RealKIE suite of five novel datasets for enterprise key information extraction. Models must convert documents to markdown and extract structured fields against a specified JSON schema. Nova 2 reports results on a human-verified version of the dataset.
Updated Aug 7, 2026
Published models3
Registry coverage3
MetricScore
EvidenceB
RealKIE-FCC leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AM Nova 2 ProAmazon | 67.0% | 100.0% | 3 | C | |
| 2 | AMNova 2 LiteAmazon | 62.1% | 50.0% | 3 | C | |
| 3 | AMNova 2 OmniAmazon | 59.8% | 0.0% | 3 | C |
Score distribution
Top published rows on the benchmark's original scale.
RealKIE-FCC highlights
The top published results on this benchmark's own scale.
What is RealKIE-FCC?
Definition and scoring fields from the benchmark registry.
RealKIE-FCC is a key information extraction benchmark drawn from real enterprise documents (FCC filings), part of the RealKIE suite of five novel datasets for enterprise key information extraction. Models must convert documents to markdown and extract structured fields against a specified JSON schema. Nova 2 reports results on a human-verified version of the dataset.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- RealKIE-FCC
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- realkie-fcc|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about RealKIE-FCC.
Which model scores highest on RealKIE-FCC?
Nova 2 Pro is currently ranked first with 67.0%.
What does RealKIE-FCC measure?
RealKIE-FCC is a key information extraction benchmark drawn from real enterprise documents (FCC filings), part of the RealKIE suite of five novel datasets for enterprise key information extraction. Models must convert documents to markdown and extract structured fields against a specified JSON schema. Nova 2 reports results on a human-verified version of the dataset.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
3 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.