M

Llama Guard 3 8B

Meta
131K
Context
$0.0200
Input /1M
$0.0600
Output /1M
N/A
Max Output
Run API Call Open in Chat

Llama Guard 3 8B

Overview

Model Name: Llama Guard 3 8B
Creator: Meta Llama
Created: February 12, 2025
Model ID: meta-llama/llama-guard-3-8b
Context Length: 131,072 tokens

Llama Guard 3 is a Llama-3.1-8B pretrained model, fine-tuned for content safety classification. The model evaluates both LLM inputs (prompt classification) and outputs (response classification), generating text that indicates whether content is safe or unsafe, plus listing violated categories when needed.

Key Capabilities

  • Content moderation across 8 languages
  • Alignment with MLCommons standardized hazards taxonomy
  • Optimized for safety and security in search and code interpreter tool calls
  • Support for Llama 3.1 capabilities

Pricing

Component Cost
Input Tokens $0.02 per 1M tokens
Output Tokens $0.06 per 1M tokens

Provider: Nebius (Token Factory)

Performance

Daily Request Volume: 26,000-43,000 requests
Token Consumption: Consistent demand across input and completion operations
Modalities: Text input / Text output

  • Llama 3.1 8B Base
  • Llama 3.1 70B Instruct
  • Llama Guard 2 7B (Previous version)

Providers

Primary Provider: Nebius (Token Factory)
API Type: OpenAI-compatible API
Data Policy: No training on user data, no prompt retention

Parameters

Supported Parameters

  • max_tokens

Technical Specifications

  • Parameter Count: 8 Billion
  • Input Modalities: Text
  • Output Modalities: Text