TensorBlock

Feedback and support: TensorBlock's Twitter/X, Telegram Group and Discord server

meta-llama/Llama-Guard-3-1B - GGUF

This repo contains GGUF format model files for meta-llama/Llama-Guard-3-1B.

The files were quantized using machines provided by TensorBlock, and they are compatible with llama.cpp as of commit b4011.

Prompt template

<|begin_of_text|><|start_header_id|>user<|end_header_id|>

Task: Check if there is unsafe content in 'User' messages in conversations according our safety policy with the below categories.

<BEGIN UNSAFE CONTENT CATEGORIES>
S1: Violent Crimes.
S2: Non-Violent Crimes.
S3: Sex Crimes.
S4: Child Exploitation.
S5: Defamation.
S6: Specialized Advice.
S7: Privacy.
S8: Intellectual Property.
S9: Indiscriminate Weapons.
S10: Hate.
S11: Self-Harm.
S12: Sexual Content.
S13: Elections.
<END UNSAFE CONTENT CATEGORIES>

<BEGIN CONVERSATION>

<END CONVERSATION>

Provide your safety assessment for ONLY THE LAST User message in the above conversation:
 - First line must read 'safe' or 'unsafe'.
 - If unsafe, a second line must include a comma-separated list of violated categories. <|eot_id|><|start_header_id|>assistant<|end_header_id|>

Model file specification

Filename Quant type File Size Description
Llama-Guard-3-1B-Q2_K.gguf Q2_K 0.667 GB smallest, significant quality loss - not recommended for most purposes
Llama-Guard-3-1B-Q3_K_S.gguf Q3_K_S 0.755 GB very small, high quality loss
Llama-Guard-3-1B-Q3_K_M.gguf Q3_K_M 0.804 GB very small, high quality loss
Llama-Guard-3-1B-Q3_K_L.gguf Q3_K_L 0.845 GB small, substantial quality loss
Llama-Guard-3-1B-Q4_0.gguf Q4_0 0.919 GB legacy; small, very high quality loss - prefer using Q3_K_M
Llama-Guard-3-1B-Q4_K_S.gguf Q4_K_S 0.923 GB small, greater quality loss
Llama-Guard-3-1B-Q4_K_M.gguf Q4_K_M 0.955 GB medium, balanced quality - recommended
Llama-Guard-3-1B-Q5_0.gguf Q5_0 1.073 GB legacy; medium, balanced quality - prefer using Q4_K_M
Llama-Guard-3-1B-Q5_K_S.gguf Q5_K_S 1.073 GB large, low quality loss - recommended
Llama-Guard-3-1B-Q5_K_M.gguf Q5_K_M 1.092 GB large, very low quality loss - recommended
Llama-Guard-3-1B-Q6_K.gguf Q6_K 1.237 GB very large, extremely low quality loss
Llama-Guard-3-1B-Q8_0.gguf Q8_0 1.600 GB very large, extremely low quality loss - not recommended

Downloading instruction

Command line

Firstly, install Huggingface Client

pip install -U "huggingface_hub[cli]"

Then, downoad the individual model file the a local directory

huggingface-cli download tensorblock/Llama-Guard-3-1B-GGUF --include "Llama-Guard-3-1B-Q2_K.gguf" --local-dir MY_LOCAL_DIR

If you wanna download multiple model files with a pattern (e.g., *Q4_K*gguf), you can try:

huggingface-cli download tensorblock/Llama-Guard-3-1B-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
Downloads last month
129
GGUF
Model size
1.5B params
Architecture
llama

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for tensorblock/Llama-Guard-3-1B-GGUF

Quantized
(15)
this model