meta-llama: Meta Llama Guard 4 12B

llama-guard-4

Context: 131k

Max Output: 1k

Input: $0.210/1M

Output: $0.210/1M

Meta’s latest safety/guardrail model for prompt and output moderation, aligning conversations to policy via classification and constrained generation.

Input: Text, Image

Output: Text

Providers

openrouter

Credits

Context131k

Max Output1k

Input (Max)$0.210/1M

Output (Max)$0.210/1M

Cache Read—

Cache Write—

groq

Credits

Context131k

Max Output1k

Input$0.200/1M

Output$0.600/1M

Cache Read—

Cache Write—

Quick Start

Use Meta Llama Guard 4 12B through Helicone's AI Gateway with automatic logging and monitoring.