Llama Guard 3 8B

Llama Guard 3 8B is a powerful Llama-3.1-8B pretrained model specifically fine-tuned for robust content safety classification. It excels at classifying content in both LLM inputs (prompt classification) and LLM responses (response classification), ensuring a safer AI interaction environment. The model generates text output indicating whether a given prompt or response is safe or unsafe, and if unsafe, it clearly lists the violated content categories. Aligned with the MLCommons standardized hazards taxonomy, Llama Guard 3 was designed to support the advanced capabilities of Llama 3.1. It provides comprehensive content moderation across 8 languages and is optimized for safety and security, including support for search and code interpreter tool calls. With a generous Context Window of 131K tokens and a Max Output of 4K tokens, it handles extensive content analysis. Access Llama Guard 3 8B for FREE on Multi AI. Pricing is competitive at $0.02 per 1M input tokens and $0.06 per 1M output tokens, making it an accessible solution for enhancing content safety in your applications.