From fb951600d75fa6ce77da43a22343cbc6ea6dfd5a Mon Sep 17 00:00:00 2001 From: Sirius Claw Date: Mon, 20 Apr 2026 16:14:27 +0000 Subject: [PATCH] chore: maximize vllm max_model_len on L4 to 32768 --- apps/base/customer1/openclaw/vllm-l4.yaml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/apps/base/customer1/openclaw/vllm-l4.yaml b/apps/base/customer1/openclaw/vllm-l4.yaml index 4e78e9a..8244462 100644 --- a/apps/base/customer1/openclaw/vllm-l4.yaml +++ b/apps/base/customer1/openclaw/vllm-l4.yaml @@ -34,8 +34,8 @@ spec: - --host=0.0.0.0 - --port=8000 - --tensor-parallel-size=1 - - --gpu-memory-utilization=0.90 - - --max-model-len=16384 + - --gpu-memory-utilization=0.95 + - --max-model-len=32768 - --enable-auto-tool-choice - --trust-remote-code - --dtype=auto