Sign in
Pending benchmarksTracking since Oct 20, 2026

Is ApplePie state of the art?Pending benchmarks.Announced Oct 20, 2026; no independent results yet.

General Data's open PII detection model built for Europe: it tags 20 kinds of personal data, including national ID formats (French NIR, German Steuer-ID, Spanish DNI, Italian codice fiscale, Dutch BSN, Polish PESEL) and IBANs, in 22 languages. Fine-tuned from XLM-RoBERTa large. Weights on Hugging Face under CC BY-NC 4.0 (free for research). Not an LLM.

Announced facts

The details so far

Lab
The General Data Corporation
Announced
Oct 20, 2026
API
Not yet
Open weights
Published
Parameters
About 559M
Type
Token classification (BIO tagging), 20 personal-data labels
Base model
FacebookAI/xlm-roberta-large
Training data
Training splits of Nemotron-PII, Gretel PII masking and MAPA, plus GD Europe: 17,950 synthetic business documents from France, Germany, Spain, Italy, the Netherlands, Portugal, Poland, Belgium, Ireland and the UK
Languages
22 listed on the card; evaluated on EU legal text in 21 languages (MAPA)
License
CC BY-NC 4.0
Hugging Face repo created
Oct 20, 2026 (UTC)

Tags marked unsure are not stated on the lab's own pages.

Next steps

What would let us call it

Verdicts use only independent boards we already track. Two consecutive snapshots are required before any crown.

  • Publicly announced with a source
  • Usable by the public (API or weights)
  • Listed on an independent board we track
  • Enough results for a rank in a category
  • Holds across two snapshots (crowns only)
Self-reported

What the lab claims

Self-reported. Shown for context; not used in any verdict.

Lab claim: 0.846 F1 on MAPA (real EU legal text, 21 languages), against 0.546 for NVIDIA GLiNER-PII, the strongest open detector it measured. Scored with the lab's own matching rule; ApplePie trained on MAPA's training split.

Receipts

Sources

LABLouai Allani (General Data) announcement on Xx.com/LouaiAllani/status/2112676338847175073 · Oct 20, 2026REPOApplePie on Hugging Facehuggingface.co/generaldata/ApplePie · Oct 20, 2026
Browse releasesRSSShareJSON

Tracking continues until an independent board lists it.