tensorblock/ALIA-40b-GGUF

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

BSC-LT/ALIA-40b - GGUF

This repo contains GGUF format model files for BSC-LT/ALIA-40b.

The files were quantized using machines provided by TensorBlock, and they are compatible with llama.cpp as of commit b4658.

Unable to determine prompt format automatically. Please check the original model repository for the correct prompt format.

Filename	Quant type	File Size	Description
ALIA-40b-Q2_K.gguf	Q2_K	15.712 GB	smallest, significant quality loss - not recommended for most purposes
ALIA-40b-Q3_K_S.gguf	Q3_K_S	18.202 GB	very small, high quality loss
ALIA-40b-Q3_K_M.gguf	Q3_K_M	20.045 GB	very small, high quality loss
ALIA-40b-Q3_K_L.gguf	Q3_K_L	21.628 GB	small, substantial quality loss
ALIA-40b-Q4_0.gguf	Q4_0	23.294 GB	legacy; small, very high quality loss - prefer using Q3_K_M
ALIA-40b-Q4_K_S.gguf	Q4_K_S	23.449 GB	small, greater quality loss
ALIA-40b-Q4_K_M.gguf	Q4_K_M	24.592 GB	medium, balanced quality - recommended
ALIA-40b-Q5_0.gguf	Q5_0	28.086 GB	legacy; medium, balanced quality - prefer using Q4_K_M
ALIA-40b-Q5_K_S.gguf	Q5_K_S	28.086 GB	large, low quality loss - recommended
ALIA-40b-Q5_K_M.gguf	Q5_K_M	28.754 GB	large, very low quality loss - recommended
ALIA-40b-Q6_K.gguf	Q6_K	33.177 GB	very large, extremely low quality loss
ALIA-40b-Q8_0.gguf	Q8_0	42.970 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/ALIA-40b-GGUF --include "ALIA-40b-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/ALIA-40b-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'