LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Meta Llama Guard 2 8B by meta-llama

 »  All LLMs  »  meta-llama  »  Meta Llama Guard 2 8B   URL Share it on

  21966   
  Arxiv:2312.06674   Arxiv:2403.13031   Autotrain compatible   Conversational   En   Endpoints compatible   Facebook   Llama   Llama-3   Meta   Pytorch   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/meta-llama/Meta-Llama-Guard-2-8B 

Meta Llama Guard 2 8B Benchmarks

LLME Score: 0.22499
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Meta Llama Guard 2 8B (meta-llama/Meta-Llama-Guard-2-8B)
🌟 Advertise your project 🚀

Meta Llama Guard 2 8B Parameters and Internals

Model Type 
text-generation, content moderation
Use Cases 
Areas:
Safety and content moderation
Applications:
Online platforms requiring content moderation
Primary Use Cases:
Classifying content for safety in both inputs and responses
Limitations:
Performance limited by training data, Not designed for chat use cases, Susceptible to adversarial attacks
Considerations:
Recommended to be used with additional solutions for unsupported categories
Additional Notes 
Supports 11 out of the 13 categories included in the MLCommons AI Safety taxonomy. Election and Defamation categories are not addressed.
Training Details 
Data Sources:
Llama Guard training set, MLCommons taxonomy, hard samples from Llama 2 70B
Methodology:
Fine-tuned for safety classification
Safety Evaluation 
Methodologies:
Harm Taxonomy, MLCommons taxonomy alignment
Findings:
Strong adaptability to other policies, Superior tradeoff between F1 score and False Positive Rate
Risk Categories:
Violent Crimes, Non-Violent Crimes, Sex-Related Crimes, Child Sexual Exploitation, Specialized Advice, Privacy, Intellectual Property, Indiscriminate Weapons, Hate, Suicide & Self-Harm, Sexual Content
Ethical Considerations:
Not disclosed
Responsible Ai Considerations 
Mitigation Strategies:
Using external components like KNN
Input Output 
Input Format:
Text prompts
Accepted Modalities:
text
Output Format:
Binary classification (safe/unsafe)
Performance Tips:
Align model with specific safety considerations for better moderation
LLM NameMeta Llama Guard 2 8B
Repository 🤗https://huggingface.co/meta-llama/Meta-Llama-Guard-2-8B 
Model Size8b
Required VRAM16.1 GB
Updated2025-09-23
Maintainermeta-llama
Model Typellama
Model Files  5.0 GB: 1-of-4   5.0 GB: 2-of-4   4.9 GB: 3-of-4   1.2 GB: 4-of-4
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licensellama3
Context Length8192
Model Max Length8192
Transformers Version4.40.0.dev0
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typebfloat16

Quantized Models of the Meta Llama Guard 2 8B

Model
Likes
Downloads
VRAM
Llama Guard Quant010913 GB
Meta Llama Guard 2 8B Bnb 4bit175 GB
Meta Llama Guard 2 8B AWQ085 GB

Best Alternatives to Meta Llama Guard 2 8B

Best Alternatives
Context / RAM
Downloads
Likes
...otron 8B UltraLong 4M Instruct4192K / 32.1 GB1531120
...a 3.1 8B UltraLong 4M Instruct4192K / 32.1 GB17624
UltraLong Thinking4192K / 16.1 GB453
...otron 8B UltraLong 2M Instruct2096K / 32.1 GB114215
...a 3.1 8B UltraLong 2M Instruct2096K / 32.1 GB8759
...otron 8B UltraLong 1M Instruct1048K / 32.1 GB714252
...a 3.1 8B UltraLong 1M Instruct1048K / 32.1 GB138729
...xis Bookwriter Llama3.1 8B Sft1048K / 16.1 GB404
Zero Llama 3.1 8B Beta61048K / 16.1 GB21
...dger Nu Llama 3.1 8B UltraLong1048K / 16.2 GB53
Note: green Score (e.g. "73.2") means that the model is better than meta-llama/Meta-Llama-Guard-2-8B.

Rank the Meta Llama Guard 2 8B Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 51545 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124
  Support LLM Explorer