org/
deepseek-r1-uncensored:latest

742 6 months ago

DeepSeek's first-generation of reasoning models with comparable performance to OpenAI-o1, including six dense models distilled from DeepSeek-R1 based on Llama and Qwen.

6 months ago

e60a9f85efb0 · 227GB

model
·
126B
·
F32
model
·
138B
·
F32
model
·
138B
·
F32
model
·
138B
·
F32
model
deepseek2
·
130B
·
Q2_K
params
{ "stop": [ "<｜begin▁of▁sentence｜>", "<｜end▁of▁sentence｜>",
template
{{- if .System }}{{ .System }}{{ end }} {{- range $i, $_ := .Messages }} {{- $last := eq (len (slice

Readme

No readme