From 169eba0734c5c3b6fed93bc0bac1517d0385076b Mon Sep 17 00:00:00 2001 From: Frank Date: Wed, 16 Sep 2026 19:51:06 -0700 Subject: [PATCH] frank-medium: checkpoints back to the build's default (0) Step 2 of the operator's six. --ctx-checkpoints 8 held per-slot KV snapshots; 0 is what llama.cpp ships, and the fleet's clients open conversations fresh. --- scripts/frank-medium.env | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/scripts/frank-medium.env b/scripts/frank-medium.env index 0d57c31..fbc64c5 100644 --- a/scripts/frank-medium.env +++ b/scripts/frank-medium.env @@ -129,7 +129,10 @@ THREAD_BATCH=8 # f16 K/V: an agentic checkpoint is the wrong place to spend KV precision. CACHE_TYPE_K=f16 CACHE_TYPE_V=f16 -CTX_CHECKPOINTS=8 +# Context checkpoints: 0 is the build's default -- no per-slot KV snapshots, so +# a re-sent or rewound conversation re-prefills instead of rewinding. Stated +# explicitly so the tunable stays visible rather than merely absent. +CTX_CHECKPOINTS=0 CACHE_RAM=8192 # Sampling defaults for requests that send none -- the values the vLLM