From cecd14123e5c415fc6794a117b4a9f4621228bad Mon Sep 17 00:00:00 2001
From: Sirius Claw <sirius.claw@siriusdevops.local>
Date: Mon, 20 Apr 2026 02:09:15 +0000
Subject: [PATCH] chore: optimize vLLM parameters for A100 by removing
 artificial limits

---
 apps/base/customer1/openclaw/vllm-gemma.yaml | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/apps/base/customer1/openclaw/vllm-gemma.yaml b/apps/base/customer1/openclaw/vllm-gemma.yaml
index 66605c7..5392bb4 100644
--- a/apps/base/customer1/openclaw/vllm-gemma.yaml
+++ b/apps/base/customer1/openclaw/vllm-gemma.yaml
@@ -38,9 +38,6 @@ spec:
             - --tensor-parallel-size=1
             - --gpu-memory-utilization=0.95        # ← Critical: lower than default 0.90
             - --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
-            - --disable-custom-all-reduce
             - --enable-auto-tool-choice 
             - --trust-remote-code
             - --dtype=auto