diff --git a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml index 4578dd1..0d058d6 100644 --- a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml +++ b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml @@ -45,7 +45,7 @@ spec: - --max-num-batched-tokens=65536 - --max-num-seqs=8 - --trust-remote-code - - --dtype=fp8 + - --dtype=auto - --enable-prefix-caching - --tool-call-parser=qwen3_xml - --reasoning-parser=qwen3