Explore Meta's LlamaGuard 2 8B Model

LlamaGuard 2 8B ensures text safety with advanced classification.

Input: text Output: text Context: 8,192 tokens Release: 2024-05-13

Meta's LlamaGuard 2 8B is a safeguard model released on May 13, 2024. Part of the Llama 3 family, it features 8 billion parameters and excels in text-to-text applications. This model focuses on classifying prompts and responses to determine their safety, identifying unsafe content and the categories violated. With a context window of 8192 tokens, LlamaGuard 2 8B supports various parameters like frequency_penalty and temperature. It is optimized for use with raw prompt input or the /completions endpoint, ensuring effective performance in text safety evaluations. The model adheres to Meta's Acceptable Use Policy.

Use Cases

Here are a few ways teams apply Meta: LlamaGuard 2 8B in practice—from fast drafting to multimodal understanding. Adapt these ideas to your workflow.

Ensure content safety in applications

Classify text for safe communication

Identify and categorize unsafe content

Enhance moderation tools

Support compliance with safety policies

Key Features

A quick look at the capabilities that make this model useful in real projects.

8 billion parameters for robust text classification

Part of the Llama 3 family

8192 token context window

Classifies prompt and response safety

Identifies unsafe content categories

Supports various tuning parameters

Specs

Overview

Vendor
meta-llama

Model ID
meta-llama/llama-guard-2-8b

Release
2024-05-13

Modalities & context

Input
text

Output
text

Context
8,192 tokens

Parameters & defaults

Supported parameters: frequency_penalty, logit_bias, max_tokens, min_p, presence_penalty, repetition_penalty, stop, temperature, top_k, top_p

Defaults: temperature 0.2, top_p 0.95

Benchmark tests: Meta: LlamaGuard 2 8B

We ran this model against a few representative prompts to show its range. Review the outputs below and be the judge.

Text

Prompt:

Write 150 words on how AI might positively upend work, leisure and creativity

in the future.

Run this prompt on Upend.AI

Ready to try Meta: LlamaGuard 2 8B?

Chat with Meta: LlamaGuard 2 8B

up.end

/ˌəpˈend/

verb

To “upend” means to completely disrupt, overturn, or drastically change the established order or structure of something. It implies a significant shift or alteration that can potentially have far-reaching consequences. When something is upended, it is turned upside down or transformed in a way that challenges conventional norms or expectations. The term often carries a sense of innovation, transformation, and sometimes even a hint of upheaval, indicating that the changes are not just minor adjustments but rather a fundamental reimagining of the status quo.

Privacy Terms