OpenGuardrails: An Open-Source Context-Aware AI Guardrails Platform

Papers with Code Trending Papers
open-source ai-safety content-safety prompt-injection jailbreaking data-leakage llm

Summary

OpenGuardrails is an open-source platform for AI safety, offering context-aware content-safety and manipulation detection (e.g., prompt injection, jailbreaking) via a unified model, plus a separate NER pipeline for data-leakage identification. It achieves state-of-the-art performance on safety benchmarks and supports private, enterprise-grade deployment.

As large language models (LLMs) become increasingly integrated into real-world applications, safeguarding them against unsafe, malicious, or privacy-violating content is critically important. We present OpenGuardrails, the first open-source project to provide both a context-aware safety and manipulation detection model and a deployable platform for comprehensive AI guardrails. OpenGuardrails protects against content-safety risks, model-manipulation attacks (e.g., prompt injection, jailbreaking, code-interpreter abuse, and the generation/execution of malicious code), and data leakage. Content-safety and model-manipulation detection are implemented by a unified large model, while data-leakage identification and redaction are performed by a separate lightweight NER pipeline (e.g., Presidio-style models or regex-based detectors). The system can be deployed as a security gateway or an API-based service, with enterprise-grade, fully private deployment options. OpenGuardrails achieves state-of-the-art (SOTA) performance on safety benchmarks, excelling in both prompt and response classification across English, Chinese, and multilingual tasks. All models are released under the Apache 2.0 license for public use.
Original Article
View Cached Full Text

Cached at: 05/25/26, 06:31 AM

Paper page - OpenGuardrails: An Open-Source Context-Aware AI Guardrails Platform

Source: https://huggingface.co/papers/2510.19169 Published on Oct 22, 2025

Abstract

OpenGuardrails is an open-source project providing a unified model for content-safety and model-manipulation detection, along with a separate NER pipeline for data-leakage identification, offering enterprise-grade deployment options.

Aslarge language models(LLMs) become increasingly integrated into real-world applications, safeguarding them against unsafe, malicious, or privacy-violating content is critically important. We presentOpenGuardrails, the first open-source project to provide both acontext-aware safetyandmanipulation detectionmodel and a deployable platform for comprehensive AI guardrails.OpenGuardrailsprotects against content-safety risks, model-manipulation attacks (e.g.,prompt injection,jailbreaking,code-interpreter abuse, and the generation/execution of malicious code), anddata leakage. Content-safety and model-manipulation detectionare implemented by aunified large model, while data-leakage identification and redaction are performed by a separatelightweight NER pipeline(e.g.,Presidio-style modelsorregex-based detectors). The system can be deployed as asecurity gatewayor anAPI-based service, with enterprise-grade, fully private deployment options.OpenGuardrailsachievesstate-of-the-art(SOTA) performance on safety benchmarks, excelling in both prompt andresponse classificationacross English, Chinese, andmultilingual tasks. All models are released under theApache 2.0 licensefor public use.

View arXiv pageView PDFProject pageGitHub344Add to collection

Get this paper in your agent:

hf papers read 2510\.19169

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper2

#### openguardrails/OpenGuardrails-Text-2510 Text Generation• 15B• UpdatedApr 9 • 1.57k • 10 #### openguardrails/OpenGuardrails-Text-4B-0124 Text Generation• 4B• UpdatedApr 9 • 103 • 1

Datasets citing this paper3

#### openguardrails/OpenGuardrailsMixZh_97k Viewer• UpdatedOct 23, 2025 • 88.5k • 213 • 3 #### qtqtqtqt/OpenGuardrailsMixZh_97k Viewer• UpdatedFeb 3 • 88.5k • 57 #### ruishen123/OpenGuardrailsMixZh_97k Viewer• UpdatedJan 14 • 88.5k • 11

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2510.19169 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Robust and Efficient Guardrails with Latent Reasoning

arXiv cs.AI

CoLaGuard is a new guardrail model that transfers multi-step safety reasoning into a continuous latent space, achieving 12.9x speedup and 22.4x token reduction compared to explicit reasoning baselines while matching macro-F1 performance on ten safety benchmarks.