From 855cce3c5403484653bd4348edaef14d588cbef5 Mon Sep 17 00:00:00 2001
From: HaimKortovich <haimkortovich88@gmail.com>
Date: Mon, 11 May 2026 11:32:56 -0500
Subject: [PATCH] fix list

---
 clusters/k3s-dgx/nim-service/qwen.yaml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/clusters/k3s-dgx/nim-service/qwen.yaml b/clusters/k3s-dgx/nim-service/qwen.yaml
index 61fae86..71c84bb 100644
--- a/clusters/k3s-dgx/nim-service/qwen.yaml
+++ b/clusters/k3s-dgx/nim-service/qwen.yaml
@@ -63,8 +63,8 @@ spec:
     - "131072"                # ↑ 32K→128K — larger decode batches = more tokens/sec
     - --max-num-seqs
     - "254"                   # ↑ Allow more concurrent sequences
-    --enable-prefix-caching
-    --dtype
+    - --enable-prefix-caching
+    - --dtype
     - "float8"                # Explicit FP8 encoding
     --quantization
     - "fbgemm-fp8"            # Explicit quantization backend