ironmagma/local-llm
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
No description
3 commits 1 branch 0 tags 42 KiB
Shell 89.2%
Python 10.8%
Find a file
Philip Peterson 01522a0573 Add the nugitea daily agent loop 
nugitea-daily.sh runs opencode with qwen3-coder over elixir_nugitea's
BACKLOG.md, one fresh session per task, and keeps a change only if it
stays in elixir_nugitea/, is under 800 lines and passes format, compile
and tests; then it commits and pushes agent/daily with its own deploy
key. opencode-agent.json denies git history, network, package and
service commands. Run by hand.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 09:23:30 -08:00
agent-loop Add the nugitea daily agent loop 2026-10-11 09:23:30 -08:00
ollama Give qwen3-coder a 128k context 2026-10-11 08:33:33 -08:00
opencode Give qwen3-coder a 128k context 2026-10-11 08:33:33 -08:00
README.md Give qwen3-coder a 128k context 2026-10-11 08:33:33 -08:00
setup.sh Ollama and opencode setup for the Mac mini 2026-10-11 07:57:46 -08:00

README.md

local-llm

Local LLM setup on the Mac mini: Ollama serving models, opencode using them.

./setup.sh

does everything, and is safe to re-run:

  • installs ollama/com.ollama.server.plist as the LaunchAgent that runs ollama serve (reloaded only when it changed, since that restarts Ollama);
  • builds each model in ollama/models/ (pulling its base model if missing);
  • symlinks ~/.config/opencode/opencode.json to opencode/opencode.json, so edits made either way end up here.

Needs brew install ollama opencode.

Models

Model From Used for
qwen3-coder:30b-128k qwen3-coder:30b opencode's default (128k context)
qwen3:32b-32k qwen3:32b opencode, as an option
inspector qwen3:32b 16k-context variant

Each derived model mostly exists to set num_ctx. Ollama's OpenAI-compatible API, which opencode talks to, ignores any context size sent with a request, so without it a long agent prompt is silently cut to Ollama's small default context. A Modelfile's # name: line is the name it's created under.

qwen3:32b is a thinking model, and in opencode it often ends its turn after thinking, without making the tool call it planned. qwen3-coder was trained for agentic tool use and doesn't think first, so it doesn't have that problem, and it's much faster: a mixture of experts with ~3B active parameters.

Server settings

From the LaunchAgent: listens on all interfaces (0.0.0.0:11434), keeps models loaded indefinitely (OLLAMA_KEEP_ALIVE=-1), flash attention on. Logs go to /tmp/ollama.log.