kyle/imladris
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

straylight: run Laguna with two slots over a unified KV pool

Browse Source 
A second request (opencode background agent, pi alongside opencode) no
longer queues behind the first. kv-unified keeps the full 131072-token
context available to either slot instead of splitting it to 65536.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
kyle
2026-09-18 00:00:27 -07:00
co-authored by Claude Opus 5
parent ab89181213
commit 14b3b3c6b8
1 changed files with 5 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
hw/straylight/default.nix
+5 -1
View File
@@ -49,7 +49,11 @@ let
spec-draft-n-max = 4
cache-type-k-draft = f16
cache-type-v-draft = f16
parallel = 1
; Two slots sharing one unified 131072-token KV pool: a second request (subagent,
; another client) runs alongside instead of queueing, and either one may still use
; the full 131072 as long as their combined context fits.
parallel = 2
kv-unified = true
ctx-size = 131072
; Previous default model (a Qwen 3.5 derivative), kept for comparison runs. It was