chore: maximize vllm max_model_len on L4 to 32768

2026-04-20 16:14:27 +00:00 · 2026-04-20 16:14:27 +00:00 · fb951600d7
commit fb951600d7
parent 399d1a58ed
1 changed files with 2 additions and 2 deletions
--- 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