From 3c542b2d4179f42cfcd0ff5dbcdb4ba28bddbd82 Mon Sep 17 00:00:00 2001 From: sirius0xdev Date: Fri, 22 May 2026 13:54:20 -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 070b1e7..5f613cf 100644 --- a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml +++ b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml @@ -41,7 +41,7 @@ spec: #- --enforce-eager - --max-model-len=136876 - --enable-auto-tool-choice - - --kv-cache-dtype=fp8 + - --kv-cache-dtype=bfloat16 - --max-num-batched-tokens=32768 - --max-num-seqs=32 - --block-size=32