fix list

2026-05-11 11:32:56 -05:00
parent 9594d6311a
commit 855cce3c54
1 changed files with 2 additions and 2 deletions
--- 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