diff --git a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml index 8b2b437..ee27362 100644 --- a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml +++ b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml @@ -39,12 +39,12 @@ spec: - --tensor-parallel-size=1 - --tokenizer-mode=hf - --gpu-memory-utilization=0.95 - - --kv-cache-dtype=fp8 + - --kv-cache-dtype=fp8_e5m2 - --max-model-len=131072 - --enable-auto-tool-choice - --enable-chunked-prefill - --max-num-batched-tokens=8192 - - --max-num-seqs=4 + - --max-num-seqs=64 - --trust-remote-code - --dtype=auto - --enable-prefix-caching