README.md

---
license: apache-2.0
base_model:
  - fastino/gliner2-privacy-filter-PII-multi
  - microsoft/mdeberta-v3-base
pipeline_tag: token-classification
tags:
  - core-ai
  - coreai
  - apple
  - on-device
  - ner
  - pii
  - gliner
  - deberta
library_name: coreai
---

# GLiNER2-PII — Core AI

The zoo's **first NER / schema-driven information-extraction** model, and its first **DeBERTa-v3**
(disentangled-attention) port. Zero-shot entity extraction: pass any label set at call time and the
model finds those entities in the text — the flagship use is **on-device PII redaction**.
[`fastino/gliner2-privacy-filter-PII-multi`](https://huggingface.co/fastino/gliner2-privacy-filter-PII-multi)
(Apache-2.0) on a multilingual [mDeBERTa-v3](https://huggingface.co/microsoft/mdeberta-v3-base) base
(278M), fused into **one static Core AI graph**; the tokenizer, schema linearization, and span
decode run in the Swift host.

> **Uncontested on iPhone.** An on-device GLiNER2 already exists (GLiNER2Swift) but it is
> macOS-CPU / MLX only. This runs the GPU on iPhone (and, AOT-compiled, the ANE) — the first GLiNER
> on Apple Silicon's accelerators.

## Use it

Three lines with [**CoreAIKit**](https://github.com/john-rocky/coreai-kit) — `InformationExtractor`
downloads this bundle once, then runs fully offline:

```swift
import CoreAIKitEmbeddings

let extractor = try await InformationExtractor(model: .gliner2PII)

// zero-shot: any labels you want, decided at call time
let entities = try await extractor.extract(
    from: "Contact Dr. Sarah Johnson at sarah.j@acme.com or +1-415-555-0142.",
    entities: ["person", "email", "phone number"])
// ["person": ["Sarah Johnson"], "email": ["sarah.j@acme.com"], "phone number": ["+1-415-555-0142"]]

// or redact in place
let clean = try await extractor.redact(
    "SSN 123-45-6789, card 4111 1111 1111 1111.",
    entities: ["social security number", "credit card number"])
// "SSN [SOCIAL SECURITY NUMBER], card [CREDIT CARD NUMBER]."
```

Runnable demo: **[Examples/InfoExtract ↗](https://github.com/john-rocky/coreai-kit/tree/main/Examples/InfoExtract)**
— a paste-text → detect-and-redact PII app (iOS + macOS CLI).

## How it works

One fused static graph runs the whole model; the host handles the text↔schema plumbing that makes it
schema-agnostic.

- **Fused graph** — `forward(input_ids[1,256], attention_mask[1,256], text_word_idx[1,96],
  schema_idx[1,17]) → span_scores[1,16,96,8]`. Inside: mDeBERTa-v3 (disentangled attention, exported
  at a fixed shape so the relative-position buckets gather cleanly) → "first" sub-word pooling →
  SpanMarker → CountLSTM → einsum → sigmoid. `MMAX=16` labels, `T=96` words, span width `K=8`.
- **Schema-agnostic** — the label set is **not** baked in. The host linearizes the user's labels
  into the graph's `input_ids` (`( [P] entities ( [E] l0 [E] l1 … ) ) [SEP_TEXT] …`) and supplies the
  gather indices, so a single converted bundle answers any schema up to `MMAX`.
- **Host collator** — mDeBERTa SentencePiece/Unigram tokenization + GLiNER word-split + schema
  linearization + first-sub-word / schema-marker gather positions. Byte-identical to GLiNER2's Python
  `collate_fn_inference`.
- **Host decode** — per-label threshold + confidence-descending greedy NMS over character spans,
  byte-identical to GLiNER2 `_format_spans`.

## Verification

Byte-gated against the reference GLiNER2 `ext.extract` at every tier — the Swift collator matches
Python `collate_fn_inference` (input ids + gather indices), the fp16 Core AI graph matches the fp32
reference (span-scores cos **0.999993**), and the decoded entities match exactly:

- **Python** — fp32 span-scores cos 1.0, decoded entities == `ext.extract`.
- **Swift on Mac GPU** — the demo PII suite decodes byte-identically; arbitrary runtime schemas
  (credentials, org/money/date/location) also match `ext.extract` exactly.
- **iPhone 17 Pro** (A19 Pro, AOT h18p) — same suite, `GATE_RESULT: PASS`. Model load ~1.8 s;
  extraction ~22–32 ms per text (warm).

## Files

- `macos/` — JIT `.aimodel` (fp16, ~582 MB) + `tokenizer/` + `extractor.json`.
- `ios/` — AOT-compiled h18p bundle (~823 MB; the device JIT is skipped) + `tokenizer/` +
  `extractor.json`.

`extractor.json` carries the graph shapes and the GLiNER special-marker token ids (they live above
the Unigram vocab, so the host emits them directly). The tokenizer is the mDeBERTa SentencePiece model
declared as `XLMRobertaTokenizer` so swift-transformers routes it through its Unigram implementation.

- Base model: [fastino/gliner2-privacy-filter-PII-multi](https://huggingface.co/fastino/gliner2-privacy-filter-PII-multi) (Apache-2.0), on [microsoft/mdeberta-v3-base](https://huggingface.co/microsoft/mdeberta-v3-base) (MIT).
- Conversion + Swift-port reference: [CoreAI-Model-Zoo](https://github.com/john-rocky/CoreAI-Model-Zoo).