tensorblock/KoSOLAR-10.9B-v1.0-GGUF

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

rrw-x2/KoSOLAR-10.9B-v1.0 - GGUF

This repo contains GGUF format model files for rrw-x2/KoSOLAR-10.9B-v1.0.

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

Prompt template

Model file specification

Filename	Quant type	File Size	Description
KoSOLAR-10.9B-v1.0-Q2_K.gguf	Q2_K	3.799 GB	smallest, significant quality loss - not recommended for most purposes
KoSOLAR-10.9B-v1.0-Q3_K_S.gguf	Q3_K_S	4.421 GB	very small, high quality loss
KoSOLAR-10.9B-v1.0-Q3_K_M.gguf	Q3_K_M	4.916 GB	very small, high quality loss
KoSOLAR-10.9B-v1.0-Q3_K_L.gguf	Q3_K_L	5.339 GB	small, substantial quality loss
KoSOLAR-10.9B-v1.0-Q4_0.gguf	Q4_0	5.740 GB	legacy; small, very high quality loss - prefer using Q3_K_M
KoSOLAR-10.9B-v1.0-Q4_K_S.gguf	Q4_K_S	5.783 GB	small, greater quality loss
KoSOLAR-10.9B-v1.0-Q4_K_M.gguf	Q4_K_M	6.103 GB	medium, balanced quality - recommended
KoSOLAR-10.9B-v1.0-Q5_0.gguf	Q5_0	6.982 GB	legacy; medium, balanced quality - prefer using Q4_K_M
KoSOLAR-10.9B-v1.0-Q5_K_S.gguf	Q5_K_S	6.982 GB	large, low quality loss - recommended
KoSOLAR-10.9B-v1.0-Q5_K_M.gguf	Q5_K_M	7.169 GB	large, very low quality loss - recommended
KoSOLAR-10.9B-v1.0-Q6_K.gguf	Q6_K	8.301 GB	very large, extremely low quality loss
KoSOLAR-10.9B-v1.0-Q8_0.gguf	Q8_0	10.751 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/KoSOLAR-10.9B-v1.0-GGUF --include "KoSOLAR-10.9B-v1.0-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/KoSOLAR-10.9B-v1.0-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'