tensorblock/falcon-7b-instruct-GGUF

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

tiiuae/falcon-7b-instruct - GGUF

This repo contains GGUF format model files for tiiuae/falcon-7b-instruct.

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

Prompt template

{system_prompt}

User: {prompt}

Assistant:

Model file specification

Filename	Quant type	File Size	Description
falcon-7b-instruct-Q2_K.gguf	Q2_K	3.595 GB	smallest, significant quality loss - not recommended for most purposes
falcon-7b-instruct-Q3_K_S.gguf	Q3_K_S	3.595 GB	very small, high quality loss
falcon-7b-instruct-Q3_K_M.gguf	Q3_K_M	3.856 GB	very small, high quality loss
falcon-7b-instruct-Q3_K_L.gguf	Q3_K_L	4.078 GB	small, substantial quality loss
falcon-7b-instruct-Q4_0.gguf	Q4_0	3.922 GB	legacy; small, very high quality loss - prefer using Q3_K_M
falcon-7b-instruct-Q4_K_S.gguf	Q4_K_S	4.420 GB	small, greater quality loss
falcon-7b-instruct-Q4_K_M.gguf	Q4_K_M	4.633 GB	medium, balanced quality - recommended
falcon-7b-instruct-Q5_0.gguf	Q5_0	4.727 GB	legacy; medium, balanced quality - prefer using Q4_K_M
falcon-7b-instruct-Q5_K_S.gguf	Q5_K_S	4.976 GB	large, low quality loss - recommended
falcon-7b-instruct-Q5_K_M.gguf	Q5_K_M	5.338 GB	large, very low quality loss - recommended
falcon-7b-instruct-Q6_K.gguf	Q6_K	6.548 GB	very large, extremely low quality loss
falcon-7b-instruct-Q8_0.gguf	Q8_0	7.145 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/falcon-7b-instruct-GGUF --include "falcon-7b-instruct-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/falcon-7b-instruct-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'