Llamacpp imatrix Quantizations of Phi 3 mini 4k instruct Using llama.cpp commit ffe6665 for quantization. Original model: https://huggingface.co/microsoft/Phi 3 mini 4k instruct All quants made using imatrix option with dataset provided by Kalomaze here Prompt format Download a file (not the whole branch) from below: Filename Quant type File Size Description Phi 3 mini 4k instruct Q8 0.gguf Q8 0 4.06GB Extremely high quality, generally unneeded but max available quant. Phi 3 mini 4k instruct Q6 K.gguf Q6 K 3.13GB Very high quality, near perfect, recommended . Phi 3 mini 4k instruct Q5 K M.gguf Q5 K M 2.81GB High quality, recommended . Phi 3 mini 4k instruct Q5 K S.gguf Q5 K S 2.64GB High quality, recommended . Phi 3 mini 4k instruct Q4 K M.gguf Q4 K M 2.39GB Good quality, uses about 4.83 bits per weight, recommended . Phi 3 mini 4k instruct Q4 K S.gguf Q4 K S 2.18GB Slightly lower quality with more space savings, recommended . Phi 3 mini 4k instruct IQ4 NL.gguf IQ4 NL 2.17GB Decent quality, slightly smaller than Q4 K S with similar performance recommended . Phi 3 mini 4k instruct IQ4 XS.gguf IQ4 XS 2.05GB Decent quality, smaller than Q4 K S with similar performance, recommended . P…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy