secfa/
DeepSeek-R1-UD-IQ2_XXS:latest

1,394 7 months ago

Unsloth's DeepSeek-R1 , I just merged the thing and uploaded it here. This is the full 671b model. MoE Bits:2.22bit Type:UD-IQ2_XXS Disk Size:183GB Accuracy:Better Details:MoE all 2.06bit. down_proj in MoE mixture of 2.5/2.06bit

... /
template
9bb403b8e0b1 · 394B
{{- if .System }}{{ .System }}{{ end }}
{{- range $i, $_ := .Messages }}
{{- $last := eq (len (slice $.Messages $i)) 1}}
{{- if eq .Role "user" }}<｜User｜>{{ .Content }}
{{- else if eq .Role "assistant" }}<｜Assistant｜>{{ .Content }}{{- if not $last }}<｜end▁of▁sentence｜>{{- end }}
{{- end }}
{{- if and $last (ne .Role "assistant") }}<｜Assistant｜>{{- end }}
{{- end }}