2 yesterday

Deimos A1 is a concise chain-of-thought (CCoT) fine-tune of Qwen3.5-4B. It produces dense, stepwise <think> blocks averaging ~1/8 the tokens of the base model while improving accuracy on every reasoning benchmark measured.

2020ccd1cc77 · 269B
{{- if .System }}<|im_start|>system
{{ .System }}<|im_end|>
{{ end -}}
{{- range .Messages }}<|im_start|>{{ .Role }}
{{ .Content }}<|im_end|>
{{ end -}}
{{- if .Prompt }}<|im_start|>user
{{ .Prompt }}<|im_end|>
{{ end -}}
<|im_start|>assistant
{{ .Response }}