tensorblock/Orion-14B-Chat-RAG-GGUF

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

OrionStarAI/Orion-14B-Chat-RAG - GGUF

This repo contains GGUF format model files for OrionStarAI/Orion-14B-Chat-RAG.

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
Orion-14B-Chat-RAG-Q2_K.gguf	Q2_K	5.508 GB	smallest, significant quality loss - not recommended for most purposes
Orion-14B-Chat-RAG-Q3_K_S.gguf	Q3_K_S	6.404 GB	very small, high quality loss
Orion-14B-Chat-RAG-Q3_K_M.gguf	Q3_K_M	7.127 GB	very small, high quality loss
Orion-14B-Chat-RAG-Q3_K_L.gguf	Q3_K_L	7.756 GB	small, substantial quality loss
Orion-14B-Chat-RAG-Q4_0.gguf	Q4_0	8.272 GB	legacy; small, very high quality loss - prefer using Q3_K_M
Orion-14B-Chat-RAG-Q4_K_S.gguf	Q4_K_S	8.334 GB	small, greater quality loss
Orion-14B-Chat-RAG-Q4_K_M.gguf	Q4_K_M	8.813 GB	medium, balanced quality - recommended
Orion-14B-Chat-RAG-Q5_0.gguf	Q5_0	10.030 GB	legacy; medium, balanced quality - prefer using Q4_K_M
Orion-14B-Chat-RAG-Q5_K_S.gguf	Q5_K_S	10.030 GB	large, low quality loss - recommended
Orion-14B-Chat-RAG-Q5_K_M.gguf	Q5_K_M	10.309 GB	large, very low quality loss - recommended
Orion-14B-Chat-RAG-Q6_K.gguf	Q6_K	11.898 GB	very large, extremely low quality loss
Orion-14B-Chat-RAG-Q8_0.gguf	Q8_0	15.409 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/Orion-14B-Chat-RAG-GGUF --include "Orion-14B-Chat-RAG-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/Orion-14B-Chat-RAG-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'