Files
microdao-daarion/models/Modelfile.qwen3.5-35b-a3b
Apple fa749fa56c chore(infra): add NODA2 setup files, docker-compose configs and root config
- AGENTS.md: Sofiia Chief AI Architect role definition
- SOFIIA_IN_OPENCODE.md, SOFIIA_NODA2_SETUP.md: NODA2 setup documentation
- agromatrix_stepan_noda1_APPLY.md, agromatrix_stepan_noda1_prod.patch: AgroMatrix production patch
- docker-compose.memory-node2.yml: memory service for NODA2
- docker-compose.node2-sofiia-supervisor.yml: sofiia supervisor for NODA2
- gateway-bot/gateway_boot.py, monitor_prompt.txt, vision_guard.py: gateway extras
- models/Modelfile.qwen3.5-35b-a3b: Qwen model definition for NODA3
- opencode.json: OpenCode providers and agents config
- scripts/init-sofiia-memory.py, scripts/node2/*, start-memory-node2.sh: NODA2 init scripts
- setup_sofiia_node2.sh: NODA2 full setup script

Made-with: Cursor
2026-03-03 07:15:20 -08:00

14 lines
432 B
Plaintext

# Qwen3.5-35B-A3B Ollama Modelfile template.
# Set MODEL_SOURCE to a GGUF path or URL before using create script.
# Example:
# export MODEL_SOURCE='hf.co/your-org/Qwen3.5-35B-A3B-GGUF:Q4_K_M'
# ./scripts/node2/install_qwen3_5_35b_a3b.sh
FROM ${MODEL_SOURCE}
PARAMETER num_ctx 8192
PARAMETER temperature 0.2
PARAMETER top_p 0.9
SYSTEM """You are Sofiia's high-capacity reasoning model. Be concise, factual, and structured."""