From 46f576dcda1163a90c0e8917de432604e50b1948 Mon Sep 17 00:00:00 2001
From: sirius0xdev <sirius0xdev@proton.me>
Date: Mon, 20 Apr 2026 02:04:04 +0000
Subject: [PATCH] dial in vllm

---
 apps/base/customer1/openclaw/vllm-gemma.yaml | 1 -
 1 file changed, 1 deletion(-)

diff --git a/apps/base/customer1/openclaw/vllm-gemma.yaml b/apps/base/customer1/openclaw/vllm-gemma.yaml
index f8622f4..66605c7 100644
--- a/apps/base/customer1/openclaw/vllm-gemma.yaml
+++ b/apps/base/customer1/openclaw/vllm-gemma.yaml
@@ -40,7 +40,6 @@ spec:
             - --max-model-len=24576                  # Start conservative (you can raise to 32768 later)
             - --max-num-batched-tokens=8192
             - --max-num-seqs=8                        # Lower for single-user / low-concurrency
-            - --enforce-eager                        # Disables CUDA graphs → big memory saver at startup
             - --disable-custom-all-reduce
             - --enable-auto-tool-choice 
             - --trust-remote-code