pplx pii masking
pplx pii masking is a 0.6B parameter bidirectional text encoder that detects and masks personally identifiable information in conversational data. It performs token level classification across 9 PII categories using a BIOES tagging scheme and produces a conversation level sensitivity score. Maximum sequence length is 4096 tokens. It is small enough to run locally, so text can be screened and sensitive spans replaced with placeholders before data is sent elsewhere. MIT licensed.
- Type
- Text
- Weights
- Closed
- Downloads
- 2.2k
- Likes
- 62
- Pipeline
- token classification
- License
- mit
Model hub
perplexity-ai/pplx-pii-masking
- Repo
- perplexity-ai/pplx-pii-masking
- Access
- Public
Technical tags and languages
More from Perplexity
10
-
Deep Research
Text
Deep Research is Perplexity’s premium reasoning assistant. It goes beyond quick Q&A by systematically searchi…
-
51k DL
Pplx Embed
Text
pplx-embed is a family of text embedding models built for real-world, web-scale retrieval.
-
5.2k DL
pplx embed context v1 0.6b
Text
A 0.6B-parameter contextual text embedding model for RAG pipelines. Unlike standard embedding models, it take…
-
3.7k DL
pplx embed context v1 4B
Text
A 4B-parameter contextual text embedding model built on diffusion-pretrained Qwen3, designed for RAG pipeline…
-
51k DL
pplx embed v1 0.6b
Text
A 0.6B parameter multilingual text embedding model built on a diffusion-pretrained Qwen3 backbone. Produces 1…
-
5.6k DL
pplx embed v1 4b
Text
A 4B-parameter text embedding model built on diffusion-pretrained Qwen3. Produces 2560-dimensional dense embe…
-
R1 1776
Text
R1 1776 is Perplexity’s open-source, post-trained variant of DeepSeek-R1 that removes CCP-aligned censorship…
-
Sonar
Text
Sonar is Perplexity’s family of reasoning models built on the R1 post-training architecture. They’re optimize…
About Perplexity
1
Perplexity provides an AI-powered answer engine that generates conversational responses supported by cited web sources.