Modelfile

# Modelfile for ThalisAI/Nanbeige4.1-3B-heretic
#
# Nanbeige4.1-3B uses Llama architecture with ChatML prompt format and
# <think>/<\/think> reasoning tokens. The default system prompt is in
# Chinese; we override it with a neutral English system prompt.
#
# Recommended parameters from the upstream model card:
#   temperature 0.6, top_p 0.95, repeat_penalty 1.0
#
# Usage:
#   ollama create nanbeige4.1-3b-heretic -f Nanbeige4.1-3B-heretic.Modelfile
#   ollama run nanbeige4.1-3b-heretic "Hello!"
#
# To use a different quantization, change the FROM line:
#   FROM hf.co/ThalisAI/Nanbeige4.1-3B-heretic:BF16

FROM hf.co/ThalisAI/Nanbeige4.1-3B-heretic:Q8_0

TEMPLATE """{{- if .System }}<|im_start|>system
{{ .System }}<|im_end|>
{{ end }}
{{- range $i, $_ := .Messages }}
{{- $last := eq (len (slice $.Messages $i)) 1 }}
{{- if eq .Role "user" }}<|im_start|>user
{{ .Content }}<|im_end|>
{{- else if eq .Role "assistant" }}<|im_start|>assistant
{{- if .Content }}
{{ .Content }}
{{- end }}{{- if not $last }}<|im_end|>{{- end }}
{{- end }}
{{- if and $last (ne .Role "assistant") }}<|im_start|>assistant
{{ end }}
{{- end }}"""

SYSTEM "You are a helpful assistant."

PARAMETER stop <|im_end|>
PARAMETER stop <|im_start|>
PARAMETER stop <|endoftext|>
PARAMETER temperature 0.6
PARAMETER top_p 0.95
PARAMETER repeat_penalty 1.0