mirror of
https://github.com/mudler/LocalAI.git
synced 2026-07-03 12:57:02 -04:00
Add a reusable current-stack paged-vs-vLLM serving snapshot harness that targets the clean DGX mirror, enforces idle/lock preflight, runs pre/post inference gates, and records ratio summaries. Assisted-by: Codex:gpt-5