openai / token-classification updated 3 months ago

privacy-filter

Name: privacy-filter
Author: openai

OpenAI Privacy Filter is a bidirectional token-classification model for personally identifiable information (PII) detection and masking in text. It is intended for high-throughput data sanitization workflows where teams need a model that they can run on-premises that is fast, context-aware, and tunable.

Params

1.4 B

Context

131,072

Downloads 30d

536 K

Likes

1,717

Commercial use: allowed apache-2.0 Not gated SAFETENSORS View on Hugging Face ↗

Download history

daily snapshots · 10 days

536 K500 K

Jul 28Jul 31Aug 3Aug 6

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
model.safetensors	bf16	5.6 GB	6.9 GB	✅ Runs comfortably
model.safetensors (bf16, full)	bf16 + 128K ctx	5.6 GB	10.0 GB	✅ Runs comfortably

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Specifications

Architecture: OpenAIPrivacyFilterForTokenClassification
Parameters: 1.4 B
Tensor type: BF16
Context length: 131,072
Vocabulary: 200,064
Layers / heads: 8 / 14
Licence: apache-2.0
First seen on the Hub: 2026-04-17
Training datasets: undisclosed
Added to our catalog: 2026-07-28

Compare with

indonesian-roberta-base-posp-tagger 3.1 M stanford-deidentifier-base 1.9 M punctuate-all 708 K ner-english-fast 543 K bert-portuguese-ner 452 K ner-english 229 K

Appears in

Best open coding models Commercial-use text models Models that fit in 12 GB