Run local inference via llama.cpp (request-scoped llama-server), drop the daemon - #36

Draft

CoreyRDean wants to merge 3 commits into

mainfrom

claude/llama-cli-local-inference-NaJIq

Commits on Jun 5, 2026

feat(local): run local inference via llama-cli, drop the server daemon
claude
committed
56444ac
feat(local): hold a request-scoped llama-server warm across the tool loop
claude
committed
76ddfe6
docs(readme): describe the request-scoped llama-server co-process
claude
committed
9aadab5