llama.cpp and ik_llama.cpp imatrix Quantizations of unsloth/phi-4-GGUF

Imatrix and quantizations created from https://huggingface.co/unsloth/phi-4-GGUF/phi-4-F16.gguf

Imatrix dataset from bartowski1182

llama.cpp

phi-4-IQ2_S.gguf
phi-4-IQ3_XS.gguf
phi-4-Q4_K_M.gguf
phi-4-IQ4_XS.gguf
phi-4-IQ4_NL.gguf

ik_llama.cpp

phi-4-IQ4_KS.gguf
phi-4-IQ4_NL_R4.gguf

Credits

llama.cpp, ik_llama.cpp, bartowski, microsoft, unsloth, huggingface

Downloads last month
34
GGUF
Model size
14.7B params
Architecture
llama
Hardware compatibility
Log In to view the estimation

2-bit

4-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for mcm07/phi-4-GGUF-imatrix

Base model

microsoft/phi-4
Quantized
(126)
this model