tensorblock/Sheared-LLaMA-1.3B-GGUF

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

princeton-nlp/Sheared-LLaMA-1.3B - GGUF

This repo contains GGUF format model files for princeton-nlp/Sheared-LLaMA-1.3B.

The files were quantized using machines provided by TensorBlock, and they are compatible with llama.cpp as of commit b4242.

Filename	Quant type	File Size	Description
Sheared-LLaMA-1.3B-Q2_K.gguf	Q2_K	0.559 GB	smallest, significant quality loss - not recommended for most purposes
Sheared-LLaMA-1.3B-Q3_K_S.gguf	Q3_K_S	0.641 GB	very small, high quality loss
Sheared-LLaMA-1.3B-Q3_K_M.gguf	Q3_K_M	0.703 GB	very small, high quality loss
Sheared-LLaMA-1.3B-Q3_K_L.gguf	Q3_K_L	0.743 GB	small, substantial quality loss
Sheared-LLaMA-1.3B-Q4_0.gguf	Q4_0	0.775 GB	legacy; small, very high quality loss - prefer using Q3_K_M
Sheared-LLaMA-1.3B-Q4_K_S.gguf	Q4_K_S	0.813 GB	small, greater quality loss
Sheared-LLaMA-1.3B-Q4_K_M.gguf	Q4_K_M	0.872 GB	medium, balanced quality - recommended
Sheared-LLaMA-1.3B-Q5_0.gguf	Q5_0	0.935 GB	legacy; medium, balanced quality - prefer using Q4_K_M
Sheared-LLaMA-1.3B-Q5_K_S.gguf	Q5_K_S	0.952 GB	large, low quality loss - recommended
Sheared-LLaMA-1.3B-Q5_K_M.gguf	Q5_K_M	1.001 GB	large, very low quality loss - recommended
Sheared-LLaMA-1.3B-Q6_K.gguf	Q6_K	1.170 GB	very large, extremely low quality loss
Sheared-LLaMA-1.3B-Q8_0.gguf	Q8_0	1.431 GB	very large, extremely low quality loss - not recommended

Firstly, install Huggingface Client

pip install -U "huggingface_hub[cli]"

Then, downoad the individual model file the a local directory

huggingface-cli download tensorblock/Sheared-LLaMA-1.3B-GGUF --include "Sheared-LLaMA-1.3B-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/Sheared-LLaMA-1.3B-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'