tensorblock/Phi-3-mini-4k-instruct-dequantized-GGUF

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

predibase/Phi-3-mini-4k-instruct-dequantized - GGUF

This repo contains GGUF format model files for predibase/Phi-3-mini-4k-instruct-dequantized.

The files were quantized using machines provided by TensorBlock, and they are compatible with llama.cpp as of commit b4242.

Run them on the TensorBlock client using your local machine ↗

Prompt template

<s><|user|>
{prompt}<|end|>
<|assistant|>

Model file specification

Filename	Quant type	File Size	Description
Phi-3-mini-4k-instruct-dequantized-Q2_K.gguf	Q2_K	1.416 GB	smallest, significant quality loss - not recommended for most purposes
Phi-3-mini-4k-instruct-dequantized-Q3_K_S.gguf	Q3_K_S	1.682 GB	very small, high quality loss
Phi-3-mini-4k-instruct-dequantized-Q3_K_M.gguf	Q3_K_M	1.955 GB	very small, high quality loss
Phi-3-mini-4k-instruct-dequantized-Q3_K_L.gguf	Q3_K_L	2.088 GB	small, substantial quality loss
Phi-3-mini-4k-instruct-dequantized-Q4_0.gguf	Q4_0	2.176 GB	legacy; small, very high quality loss - prefer using Q3_K_M
Phi-3-mini-4k-instruct-dequantized-Q4_K_S.gguf	Q4_K_S	2.189 GB	small, greater quality loss
Phi-3-mini-4k-instruct-dequantized-Q4_K_M.gguf	Q4_K_M	2.393 GB	medium, balanced quality - recommended
Phi-3-mini-4k-instruct-dequantized-Q5_0.gguf	Q5_0	2.641 GB	legacy; medium, balanced quality - prefer using Q4_K_M
Phi-3-mini-4k-instruct-dequantized-Q5_K_S.gguf	Q5_K_S	2.641 GB	large, low quality loss - recommended
Phi-3-mini-4k-instruct-dequantized-Q5_K_M.gguf	Q5_K_M	2.815 GB	large, very low quality loss - recommended
Phi-3-mini-4k-instruct-dequantized-Q6_K.gguf	Q6_K	3.136 GB	very large, extremely low quality loss
Phi-3-mini-4k-instruct-dequantized-Q8_0.gguf	Q8_0	4.061 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/Phi-3-mini-4k-instruct-dequantized-GGUF --include "Phi-3-mini-4k-instruct-dequantized-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/Phi-3-mini-4k-instruct-dequantized-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'