akuldatta / deepseek-coder-v2-lite

16B

63 Pulls Updated 6 weeks ago

2 months ago

1fd1940effe9 · 8.6GB

model
deepseek2
·
15.7B
·
IQ1_M
template
{{ if .System }}System: {{ .System }} {{ end }}{{ if .Prompt }}User: {{ .Prompt }} {{ end }}Assistant: <|begin_of_text|>{{ .Response }}

Readme

Models:
DeepSeek Coder V2 Instruct Lite Quantized with IQ4_XS (https://ollama.com/akuldatta/deepseek-coder-v2-lite) and Q5_K_S (https://ollama.com/akuldatta/deepseek-coder-v2-lite:q5ks) quants.