PI

PII Tracer

PiiGGUF by perplexity-aiUpdated 2026-09-22
Download on Hugging Face
Parameters596M
Context32K
Layers28
Architectureqwen3
Smallest quant1.2 GB
Quick answer

PII Tracer needs about 2.6 GB of memory at F16 (a 1.2 GB file) with an 8K context window, so it fits on a 4 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 3.8 GB.

PII Tracer – a 600 M‑parameter Qwen3 encoder for PII detection

PII Tracer is a 0.596 B parameter bidirectional Qwen3 encoder with two heads: a token classification head for 9 PII categories and a sensitivity head for conversation‑level classification. The GGUF version uses an f16 backbone and fp32 heads, and is served via llama‑cpp.

Good at

  • Detects 9 PII categories with BIOES tagging
  • Classifies conversation sensitivity
  • Runs on llama‑cpp with f16 backbone

Limits

  • Requires 8 k context for full model
  • No vision support
  • MIT license only
  • Limited to conversational data

Best for

  • PII masking in chat logs
  • Privacy compliance checks
  • Sensitive data extraction

Who it’s for. Developers building privacy‑aware chat applications or data‑processing pipelines that need automated PII detection. The model is lightweight enough for local inference with a modest VRAM budget.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
F161.2 GB 2.6 GB
Fits Get file →

The only available quant is F16, 1.2 GB file size. It requires about 2.6 GB VRAM at 8 k context, which is approximate. For lower VRAM, no other quant is provided.

What GPU do you need for PII Tracer?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is F16 (1.2 GB file, about 2.6 GB in use).

Which PII Tracer quant should you choose?

F16 (1.2 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen3
Parameters596M
Layers28
Trained context32,768 tokens
KV cache (fp16)112 KB per token

Source

Quantized byperplexity-ai
FamilyPii
Licensemit
Downloads (HF)16,211
Last updated2026-09-22

Frequently asked questions about PII Tracer

How much VRAM does PII Tracer need?

About 2.6 GB at F16 with an 8K context window. That is the 1.2 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run PII Tracer on an 8 GB GPU?

Yes, with F16: the 1.2 GB file needs about 2.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of PII Tracer GGUF?

It depends on the quant: F16 is 1.2 GB.

What context length does PII Tracer support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 0.9 GB at 8K and 3.8 GB at 32K.

Where can I download PII Tracer GGUF?

The files are on Hugging Face in the perplexity-ai/PII-Tracer-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does PII Tracer use?

Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run PII Tracer on 8 GB VRAM?

Yes, the F16 quant uses roughly 2.6 GB VRAM at 8 k context, leaving headroom for other processes.

Does PII Tracer support image inputs?

No, the model is designed for text only and has no vision capabilities.

Similar models

Browse all →

More from perplexity-ai

All by perplexity-ai →

More Pii models

All Pii →

Other tiny models

All tiny models →