From 99c7ec531f626bc16a2ee24fe19eb841b4ee8d32 Mon Sep 17 00:00:00 2001
From: Frank <frank@dankware>
Date: Thu, 20 Aug 2026 17:09:23 -0700
Subject: [PATCH] frank-default-small: ncpu-moe=32(8GPU, ~51.1 t/s gen)

---
 scripts/frank-default-small.env | 4 ++--
 scripts/serve-frank-small.sh    | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/scripts/frank-default-small.env b/scripts/frank-default-small.env
index 1ed0802..117bffa 100644
--- a/scripts/frank-default-small.env
+++ b/scripts/frank-default-small.env
@@ -31,8 +31,8 @@ NGRAM_MOD_N_MIN=48
 NGRAM_MOD_N_MAX=64
 NGRAM_MOD_N_MATCH=24
 
-# Expert offload: 35 of 40 layers keep MoE weights on CPU; 5 on GPU.
-N_CPU_MOE=35
+# Expert offload: 32 of 40 layers keep MoE weights on CPU; 8 on GPU.
+N_CPU_MOE=32
 N_GPU_LAYERS=all
 
 # KV cache on GPU, both K and V at f16.
diff --git a/scripts/serve-frank-small.sh b/scripts/serve-frank-small.sh
index b6325f0..909303e 100755
--- a/scripts/serve-frank-small.sh
+++ b/scripts/serve-frank-small.sh
@@ -36,7 +36,7 @@ fi
 : "${CACHE_TYPE_K:=f16}"
 : "${CACHE_TYPE_V:=f16}"
 : "${VERBOSITY:=4}"
-: "${N_CPU_MOE:=35}"
+: "${N_CPU_MOE:=32}"
 : "${N_GPU_LAYERS:=all}"
 : "${BATCH_SIZE:=4096}"
 : "${UBATCH_SIZE:=2048}"