From: Anderson A. Giacomolli <agiacomolli@gmail.com>
Date: Sat, 7 Nov 2026 01:54:20 +0000 (-0300)
Subject: Remove the warm-up request after the turns
X-Git-Url: http://git.giacomolli.uk/onechat/commitdiff_plain/f59bb67ae1bc2c899ef4f6be9ce2ff6296c7ffcc?ds=sidebyside;hp=f59bb67ae1bc2c899ef4f6be9ce2ff6296c7ffcc

Remove the warm-up request after the turns

The first request of a turn reads the view of the turn before from the
prompt cache without a warm-up. In two live runs of 6 turns on a view of
about 31,000 tokens, turns 2 to 6 read 99.1% to 99.4% of their input
from the cache without the warm-up, and 99.3% to 99.6% with it. One call
per turn for about 0.1 points is not worth it, so the loop sends no
warm-up and the client does not stop a stream early.

The view messages that only grow stay. With the whole view in one user
message, turn 2 of a third run read nothing from the cache, and turns 3
to 6 read 99.2% to 99.5%. With the grow-only messages, each of the 10
later turns hit.

Issue: 5daf23e
---