Modelfile-llama3.1-functionary

#  ollama create sammcj/functionary-small-v3.2:q6_k -f modelfiles/Modelfile-llama3.1-functionary

FROM ../functionary-small-v3.2-q6_k.gguf

#  Modified llama 3.1 template based on https://github.com/MeetKai/functionary/blob/main/functionary/prompt_template/llama31_prompt_template.py
TEMPLATE """{{- if or .System .Tools }}<|start_header_id|>system<|end_header_id|>
{{- if .System }}

{{ .System }}
{{- end }}
{{- if .Tools }}

When you receive a tool call response, use the output to format an answer to the original user question.

You are a helpful assistant with tool calling capabilities. You call functions using the format <function=name>arguments</function>

Available functions:
{{- range $.Tools }}
{{- . }}
{{ end }}
{{- end }}<|eot_id|>
{{- end }}
{{- range $i, $_ := .Messages }}
{{- $last := eq (len (slice $.Messages $i)) 1 }}
{{- if eq .Role "user" }}<|start_header_id|>user<|end_header_id|>

{{ .Content }}<|eot_id|>{{ if $last }}<|start_header_id|>assistant<|end_header_id|>

{{ end }}
{{- else if eq .Role "assistant" }}<|start_header_id|>assistant<|end_header_id|>
{{- if .ToolCalls }}
{{- range .ToolCalls }}
<function={{ .Function.Name }}>{{ .Function.Arguments }}</function>
{{- end }}
{{- else }}

{{ .Content }}
{{- end }}{{ if not $last }}<|eot_id|>{{ end }}
{{- else if eq .Role "tool" }}
<|python_tag|>{{ .Content }}<|eot_id|>{{ if $last }}<|start_header_id|>assistant<|end_header_id|>

{{ end }}
{{- end }}
{{- end }}"""

PARAMETER stop "<|start_header_id|>"
PARAMETER stop "<|end_header_id|>"
PARAMETER stop "<|eot_id|>"
PARAMETER stop "<|end_of_text|>"
PARAMETER stop "<|eom_id|>"
PARAMETER stop "<|reserved_special_token"


# 8k is small but hopefully the function calls are small as well.
PARAMETER num_ctx 8192

## For codegen ##
#  PARAMETER temperature 0.3
#  PARAMETER top_p 0.8

#  PARAMETER presence_penalty 0.2
#  PARAMETER frequency_penalty 0.2
#  PARAMETER repeat_last_n 50


### min_p sampling ##
# min_p works best with a bit of temperature
PARAMETER temperature 0.2
# 1.0 disables top_p, so we can use min_p
PARAMETER top_p 1.0
PARAMETER min_p 0.9
### min_p sampling ##