tensorblock/Falcon2-5.5B-Polish-GGUF

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

ssmits/Falcon2-5.5B-Polish - GGUF

This repo contains GGUF format model files for ssmits/Falcon2-5.5B-Polish.

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

System: {system_prompt}
User: 
{prompt}
Falcon:

Filename	Quant type	File Size	Description
Falcon2-5.5B-Polish-Q2_K.gguf	Q2_K	2.033 GB	smallest, significant quality loss - not recommended for most purposes
Falcon2-5.5B-Polish-Q3_K_S.gguf	Q3_K_S	2.347 GB	very small, high quality loss
Falcon2-5.5B-Polish-Q3_K_M.gguf	Q3_K_M	2.558 GB	very small, high quality loss
Falcon2-5.5B-Polish-Q3_K_L.gguf	Q3_K_L	2.725 GB	small, substantial quality loss
Falcon2-5.5B-Polish-Q4_0.gguf	Q4_0	2.990 GB	legacy; small, very high quality loss - prefer using Q3_K_M
Falcon2-5.5B-Polish-Q4_K_S.gguf	Q4_K_S	2.990 GB	small, greater quality loss
Falcon2-5.5B-Polish-Q4_K_M.gguf	Q4_K_M	3.190 GB	medium, balanced quality - recommended
Falcon2-5.5B-Polish-Q5_0.gguf	Q5_0	3.595 GB	legacy; medium, balanced quality - prefer using Q4_K_M
Falcon2-5.5B-Polish-Q5_K_S.gguf	Q5_K_S	3.595 GB	large, low quality loss - recommended
Falcon2-5.5B-Polish-Q5_K_M.gguf	Q5_K_M	3.799 GB	large, very low quality loss - recommended
Falcon2-5.5B-Polish-Q6_K.gguf	Q6_K	4.238 GB	very large, extremely low quality loss
Falcon2-5.5B-Polish-Q8_0.gguf	Q8_0	5.411 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/Falcon2-5.5B-Polish-GGUF --include "Falcon2-5.5B-Polish-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/Falcon2-5.5B-Polish-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'