ModernBERT PromptGuard is a high-performance binary classifier specifically designed to detect malicious prompts in large language model applications, including prompt injection and jailbreak attacks.

Natural Language ProcessingTransformersEnglish

Granite Guardian 3.2 5B is an AI risk detection model developed by IBM, specifically designed to detect various security risks in prompts and responses. Based on the IBM AI Risk Atlas, this model can identify multiple risk types such as harm, social bias, jailbreaking, and violence, and is an important tool for enterprise-level AI security monitoring.

Natural Language ProcessingTransformers

Llama Prompt Guard 2 is a series of prompt attack detection models launched by Meta, including an upgraded 86M-parameter version and a lightweight 22M-parameter version, designed to detect prompt injection and jailbreak attacks in large language model applications.

Natural Language ProcessingTransformersMultiple Languages

Llama Prompt Guard 2 86M is a text classification model designed to detect prompt injection and jailbreak attacks, serving as the second-generation product in the Prompt Guard series.

Natural Language ProcessingTransformersMultiple Languages

Prompt Guard is a text classification model designed to detect prompt attacks, capable of identifying malicious prompt injections and jailbreak attempts.

Natural Language ProcessingSafetensors

Labess-7b-chat is an open-source model specifically instruction-tuned for the Tunisian dialect (Derja), based on the jais-adapted-7b-chat model and continuously pre-trained using Tunisian dialect datasets.

Natural Language ProcessingTransformersArabic

TestSavantAI models are a set of fine-tuned classifiers specifically designed to defend against prompt injection and jailbreak attacks targeting large language models (LLMs).

Natural Language ProcessingTransformersEnglish

The TestSavantAI model is a set of classifiers specifically designed to defend against prompt injection and jailbreak attacks in large language models (LLMs). The tiny version is based on the BERT-tiny architecture, balancing security and computational efficiency.

Natural Language ProcessingTransformersEnglish

A small BERT model for detecting saturation-type jailbreak attacks, not suitable for independently defending against other types of jailbreak attacks.

Natural Language ProcessingTransformersEnglish

A security protection model fine-tuned based on DeBERTa-v3-large, used to detect unsafe content in conversations with large language models and prevent jailbreak attacks.

Natural Language ProcessingTransformers

Atlas-Chat is a series of open-source large language models specifically fine-tuned for the Moroccan spoken language Darija. It is part of the JAIS project. The model has 2 billion parameters and can efficiently generate fluent Moroccan Darija text, suitable for deployment in resource-constrained environments.

Natural Language ProcessingTransformersMultiple Languages

The Jais series is a bilingual large language model optimized for Arabic, while also possessing strong English capabilities. The 30B-16K version has 30 billion parameters and supports a context length of 16,384 tokens.

Natural Language ProcessingSafetensorsMultiple Languages

The Jais series are large language models specialized in Arabic processing while possessing strong English capabilities, with variants ranging from 590 million to 70 billion parameters.

Natural Language ProcessingPytorchMultiple Languages

The Jais series is a comprehensive English-Arabic bilingual large language model, optimized for Arabic while maintaining strong English capabilities. This model has been fine-tuned for instruction following, making it suitable for conversational scenarios.

Natural Language ProcessingSafetensorsMultiple Languages

The Jais series is a large language model optimized for Arabic and bilingual in English and Arabic. The 13B version has 13 billion parameters and supports a context length of 2048 tokens.

Natural Language ProcessingSafetensorsMultiple Languages

The Jais series is a large English-Arabic bilingual language model specifically optimized for Arabic, with strong English capabilities and 670 million parameters

Natural Language ProcessingSafetensorsMultiple Languages

The Jais family is a series of large bilingual language models specifically optimized for Arabic while maintaining strong English capabilities. This 6.7B parameter model is pre-trained from scratch and supports Arabic and English text generation.

Natural Language ProcessingSafetensorsMultiple Languages

The Jais series is a bilingual large language model specifically optimized for Arabic and English, featuring variants ranging from 590 million to 70 billion parameters.

Natural Language ProcessingPytorchMultiple Languages

Jais is a bilingual large language model family specifically optimized for Arabic, with strong English capabilities, ranging from 590 million to 70 billion parameters

Natural Language ProcessingSafetensorsMultiple Languages

Jais Adapted 13B is a bilingual (Arabic-English) large language model based on the Llama-2 architecture, enhanced with adaptive pretraining for Arabic language capabilities

Natural Language ProcessingSafetensorsMultiple Languages