---
license: apache-2.0
base_model: openai/privacy-filter
tags:
  - mlx
  - token-classification
  - privacy
  - pii-detection
  - bioes
library_name: mlx
pipeline_tag: token-classification
---

# privacy-filter-mlx (int4)

MLX-converted, int4-quantized weights of [openai/privacy-filter](https://huggingface.co/openai/privacy-filter),
packaged for use with [PrivacyFilterKit](https://github.com/kokluch/privacy-filter-swift) — a Swift package
that runs on-device PII detection on Apple platforms via [MLX-Swift](https://github.com/ml-explore/mlx-swift).

## Bundle contents

| File | Purpose |
|------|---------|
| `weights.safetensors` | int4 affine-quantized weights (group_size=64). Embedding + classifier head kept full-precision. |
| `tokenizer.json` | Hugging Face tokenizer (copied verbatim from upstream). |
| `tokenizer_config.json` | Tokenizer config. |
| `id2label.json` | 33-label BIOES table (8 entity types: account_number, private_address, private_date, private_email, private_person, private_phone, private_url, secret). |
| `model_config.json` | Architecture parameters consumed by the Swift runtime. |
| `MANIFEST.json` | SHA-256 hashes of every file in the bundle. |

## Architecture

- 8 transformer layers, hidden size 640, 14 attention heads (2 KV heads, GQA)
- 128 local experts, top-4 MoE routing
- 200 064 vocab, 131 072 max position embeddings, sliding-window attention (128)
- 33-label BIOES head; the Swift decoder derives a BIOES validity mask at runtime
  (no learned CRF transition matrix in the upstream checkpoint)

## Usage (Swift)

```swift
import PrivacyFilterKit

let bundle = URL(fileURLWithPath: "/path/to/privacy-filter-int4-v0.1.0")
let filter = try await PrivacyFilter(source: .directory(bundle))
let entities = try await filter.detect(in: "Email me at jane@example.com")
```

See the [PrivacyFilterKit README](https://github.com/kokluch/privacy-filter-swift) for the full API.

## Conversion pipeline

The conversion was produced by the scripts in [`privacy-filter-swift/scripts/`](https://github.com/kokluch/privacy-filter-swift/tree/main/scripts):

1. `01_download_hf.py` — download the upstream checkpoint
2. `02_export_config.py` — extract label table, tokenizer, normalized model config
3. `03_convert_mlx.py` — rename keys, downcast to bf16, write MLX-friendly safetensors
4. `04_quantize_mlx.py` — int4 affine quantization (embedding + classifier head full-precision)
5. `06_export_bundle.py` — assemble bundle + MANIFEST + tar.gz archive

## License

Apache 2.0, inherited from the upstream model. See [LICENSE](https://www.apache.org/licenses/LICENSE-2.0.txt).

## Modifications from upstream

This bundle is a derivative of `openai/privacy-filter`. Significant changes:

- Weights converted from PyTorch safetensors to MLX-format safetensors (key rename + bf16 cast).
- int4 affine-quantized (group_size=64). Embedding, classifier head, and any transition matrix
  are kept full-precision.
- Bundle adds `model_config.json`, `id2label.json`, and `MANIFEST.json` for the Swift runtime;
  no model logic is changed.

## Credits

- Upstream model: [`openai/privacy-filter`](https://huggingface.co/openai/privacy-filter)
- Swift runtime: [PrivacyFilterKit](https://github.com/kokluch/privacy-filter-swift)
- Conversion runtime: [MLX](https://github.com/ml-explore/mlx) / [MLX-Swift](https://github.com/ml-explore/mlx-swift)