diff --git a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml index 5f613cf..186d1d6 100644 --- a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml +++ b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml @@ -39,7 +39,7 @@ spec: - --tokenizer-mode=hf - --gpu-memory-utilization=0.90 #- --enforce-eager - - --max-model-len=136876 + - --max-model-len=131072 - --enable-auto-tool-choice - --kv-cache-dtype=bfloat16 - --max-num-batched-tokens=32768