llamaguard
orchestra-research889 installs
Filter unsafe prompts and replies with LlamaGuard.
What it does
Guides integration of LlamaGuard checks that filter unsafe prompts and responses in LLM applications.
When to use it
Use when building production LLM applications that need a pretrained moderation model with detailed safety categories.
How to use it
Give it an LLM application to protect; it supplies code for LlamaGuard prompt and response filtering.
What you provide
- An existing project
Uses
Latest review
What you need · 2
The moderation examples run as Python code; no Python version is specified.
Requires signing in to HuggingFace; the access-denied instructions also require accepting the model license.
Access · 3
This skill
HuggingFace
Read
Downloads LlamaGuard model files
transformers
Execute
Installs transformers