Best-in-Class
PII Dectection,
Open Source


A 300M-parameter multilingual model for detecting
and redacting personally identifiable information.
Built for production privacy workflows.
0.471
avg. SPY
F1 score
42
entity
types
300M
params
encoder
RESEARCH
Capabilities of GLiNER2-PII

PII extraction. Identify and extract 42 entity types across seven categories - 5× coverage of OpenAI's Privacy Filter.
Learn more

Customizable schema at inference. The target schema is an input to the model, not baked into the weights.
Learn more

One forward pass, deterministic inference. Model evaluates all 42 entity types simultaneously.
Deploy GLiNER2-PII
EFFICIENCY
Smaller, faster, built to self-host
SOTA performance at fraction of the size
vs OpenAI Privacy Filter.
Low latency
Optimized for real-time, production pipelines.
Stay on your infrastructure
Apache 2.0 license. Can run on-prem or air-gapped, or deployed with Pioneer.
< 100ms
Average latency
per request
5x
Label coverage
versus OpenAI Privacy Filter
300M
Parameters
versus OpenAI's 1.5B
1,100,000+
Monthly downloads
of GLiNER model family

GLiNER2-PII Benchmark Results
Evaluated on SPY (Synthetic PII Yesterday), 200 documents split evenly between legal Q&A forums and medical transcripts. None of the models in the comparison were trained on this data, making it an out-of-distribution test.
OpenAI Privacy Filter
NVIDIA GLiNER PII
FASTINO GLiNER2-PII
Parameters
1.5B
—
0.3B
Entity types
8 (fixed)
—
42 (customizable)
Architecture
Decoder (autoregressive)
Encoder (GLiNER)
Encoder (GLiNER2)
Custom schema at inference
Recall - Legal domain
0.640
0.431
0.722
Precision - Legal domain
0.250
0.374
0.522
F1 score - Legal domain
0.360
0.401
0.475
Recall - Medical domain
0.671
0.431
0.681
Precision - Medical domain
0.271
0.341
0.355
F1 score - Medical domain
0.386
0.381
0.467
1,100,000+
monthly
downloads
2400+
github
stars
1.1BN+
end
users
Join the community
Join our active community on discord
Join now
Need help?
Get in touch with our support team.
Contact Support
Fastino Inc. (“Fastino”) develops specialized AI models and provides APIs designed to support structured data extraction, classification, reasoning, and production AI workflows. Fastino is a technology company and does not provide legal, financial, compliance, or advisory services.
Any outputs, predictions, classifications, or decisions generated through Fastino models are based on the configuration, data, and implementation provided by the customer. Fastino does not control, verify, or guarantee the accuracy, completeness, or suitability of model outputs for any specific purpose. By using this website or Fastino’s models and services, you acknowledge that all content and outputs are provided for informational and operational purposes only and agree to our Terms of Use and Privacy Policy.
2026 Fastino Inc.
All rights reserved