diff --git a/apps/base/customer1/openclaw/vllm-l4.yaml b/apps/base/customer1/openclaw/vllm-l4.yaml index b630370..a1c8013 100644 --- a/apps/base/customer1/openclaw/vllm-l4.yaml +++ b/apps/base/customer1/openclaw/vllm-l4.yaml @@ -38,7 +38,7 @@ spec: - --port=8000 - --tensor-parallel-size=1 - --gpu-memory-utilization=0.95 - - --max-model-len=24576 + - --max-model-len=32768 - --kv-cache-dtype=fp8 - --max-num-batched-tokens=2048 - --max-num-seqs=128