From fa6dd3c8a6cede3eee4b8a60eb2eb9cac75a5bb2 Mon Sep 17 00:00:00 2001 From: sirius0xdev Date: Fri, 22 May 2026 23:39:27 -0400 Subject: [PATCH] Update rtx6000-vllm.yaml --- infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml index 5cd1928..2baa422 100644 --- a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml +++ b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml @@ -42,7 +42,7 @@ spec: - --enable-auto-tool-choice - --kv-cache-dtype=bfloat16 - --max-num-batched-tokens=65536 - - --max-num-seqs=64 + - --max-num-seqs=32 - --block-size=32 - --language-model-only - --enable-chunked-prefill