Phi-3 128K version of Q5
12.4K Pulls Updated 7 months ago
Updated 7 months ago
7 months ago
d8f0cff3daa1 · 2.7GB
model
archllama
·
parameters3.82B
·
quantizationQ5_K_M
2.7GB
params
{"stop":["\u003c|user|\u003e","\u003c|assistant|\u003e","\u003c|system|\u003e","\u003c|end|\u003e","
127B
template
{{ if .System }}<|system|>
{{ .System }}<|end|>
{{ end }}{{ if .Prompt }}<|user|>
{{ .Prompt }}<|end
149B
Readme
Convert from PrunaAI/Phi-3-mini-128k-instruct-GGUF-Imatrix-smashed, adopt Q5_K_8_4 quantization.
Its multilingual capabilities are clearly superior to version Q4 quantization.