RedHatAI/Llama-Guard-4-12B-quantized.w4a16 is a 4-bit quantized version of the Llama Guard 4 12B safety model. It classifies user and assistant messages into categories such as violent crimes, sex crimes, child exploitation, defamation, and others. The model is designed for integration into LLM deployment pipelines to enforce responsible AI usage.
In the Foundation models & chat space, Llama Guard 4 12B Quantized.w4a16 takes a focused approach. It focuses on detecting and classifying unsafe or harmful content across multiple risk categories in LLM inputs and outputs. Llama Guard 4 12B Quantized.w4a16 is an open-source project aimed at developers. Llama Guard 4 12B Quantized.w4a16 is open source under the EUPL-1.2 license. It ships for the web and API.
RedHatAI builds and maintains Llama Guard 4 12B Quantized.w4a16, and it first shipped in 2024. Among its 3 catalogued features are Content Moderation, Safety Classification, and Quantized Model.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do