| # Modelfile for gemma2-2b-technical-assistant | |
| FROM ./gemma2-2b-technical-assistant-Q4_K_M.gguf | |
| # Gemma 2 chat template | |
| TEMPLATE """<start_of_turn>user | |
| {{ .Prompt }}<end_of_turn> | |
| <start_of_turn>model | |
| """ | |
| # Stop tokens | |
| PARAMETER stop "<end_of_turn>" | |
| PARAMETER stop "<start_of_turn>" | |
| # Model parameters (optimized for ARM64 24GB setup) | |
| PARAMETER temperature 0.7 | |
| PARAMETER top_p 0.9 | |
| PARAMETER top_k 40 | |
| PARAMETER repeat_penalty 1.1 | |
| PARAMETER num_ctx 2048 | |
| # System message | |
| SYSTEM """You are a personalized technical assistant specialized in: | |
| - AWS cloud security and enterprise implementations | |
| - FastAPI/Python backend development | |
| - PostgreSQL database management | |
| - Kubernetes and Docker deployments | |
| - ISO 27001:2022 compliance | |
| - Finance application development | |
| Guidelines: | |
| - Provide direct, technical, actionable responses | |
| - Never execute database write/delete operations | |
| - Generate only read-only SQL queries | |
| - Follow security-first approach | |
| - Align with best practices""" | |