Task
Extraction of every field, including line items on delivery notes and invoices.
Classification and more document types will follow.
How reliably do language models extract the data from your documents? Measured with Lector: the same documents, the same expected values, every field checked individually.
Leaderboard
Ranking by field accuracy …
As of …| Rank | Model | |
|---|---|---|
| Loading ranking … | ||
Field accuracy = correct ÷ expected fields. More under methodology.
Document examples
Generated delivery notes, invoices and claim notifications from the test set with expected values and the models’ actual responses. A selection, not the full dataset. Document content and field names remain in German.
Document
Model result
Header data
Line items
A field is correct if its value matches the expected value exactly. A missing value counts as an error.
Methodology
Every model extracts the same documents in Lector.
Each field is compared with its expected value.
Extraction of every field, including line items on delivery notes and invoices.
Classification and more document types will follow.
Generated documents with expected values:
Field accuracy = correct ÷ expected fields. Missing, incorrect and invented values count as errors.
Values must match exactly; the order of line items doesn’t matter.
Evaluation CenterBeta
The leaderboard above comes from Lector’s Evaluation Center. There you measure models, prompts and settings on your own documents, with the same metrics.
Verified documents with expected values, for example everything marked as gold.
Each variant, whether a different model or a revised prompt, runs on its own branch against the same dataset.
Field accuracy per variant, field and document type side by side. The best variant goes live.