FROM ./cve-cwe-gemma4-12b-Q8_0.gguf

# Gemma 4 turn format with thinking OFF (empty thought channel).
TEMPLATE """{{ if .System }}<|turn>system
{{ .System }}<turn|>
{{ end }}<|turn>user
{{ .Prompt }}<turn|>
<|turn>model
<|channel>thought
<channel|>{{ .Response }}<turn|>
"""

SYSTEM """You are a vulnerability analyst. Given a CVE description, reply with only the CWE ID(s) it maps to, comma-separated."""

PARAMETER temperature 0
PARAMETER stop "<turn|>"
PARAMETER stop "<|turn>"
# Ollama runs gemma4 with thinking ON by default (wastes ~100-900 tokens/query).
# Pass "think": false (API) or run `/set nothink` (interactive) for fast output.