From 18aa77923240c42592dafa81b17b27f82015ea37 Mon Sep 17 00:00:00 2001 From: sirius0xdev Date: Sat, 16 May 2026 23:01:39 -0400 Subject: [PATCH] Update rtx6000-vllm.yaml --- infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml index d95af39..abaab2b 100644 --- a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml +++ b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml @@ -51,7 +51,8 @@ spec: - --enable-prefix-caching - --tool-call-parser=qwen3_xml - --reasoning-parser=qwen3 - - --speculative-config "{"method": "mtp", "num_speculative_tokens": 2}" + - --speculative-config + - --'{"method": "mtp", "num_speculative_tokens": 2}' ports: - containerPort: 8000