Back to vendors

AI security vendor

Meta Platforms, Inc.

Commercial parent organization

Official website

2

Offerings

2

Reviewed claims

1

Capabilities

Domains

Runtime Protection

Actions

DETECT

Offerings

Open an offering for reviewed behavior and evidence

Meta Platforms, Inc. · project

Llama Guard

1 claims+

Safeguard model that classifies prompts and model responses against content-safety taxonomies.

Official source
Prompt and Response Guardrails
DETECT

Llama Guard classifies input prompts and model responses against content-safety categories so applications can detect unsafe interactions.

Meta Platforms, Inc. · project

Llama Prompt Guard

1 claims+

Classifier models for detecting prompt injection and jailbreak inputs before they reach an application model.

Official source
Prompt and Response Guardrails
DETECT

Llama Prompt Guard classifies input text for prompt injection and jailbreak patterns before the text reaches the primary model.