Update rtx6000-vllm.yaml

2026-05-04 21:59:35 -04:00 · 2026-05-04 21:59:35 -04:00 · 16c8ebc41b
commit 16c8ebc41b
parent e0971380ad
1 changed files with 1 additions and 1 deletions
--- a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml
+++ b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml
@ -45,7 +45,7 @@ spec:
            - --max-num-batched-tokens=65536
            - --max-num-seqs=8
            - --trust-remote-code
-            - --dtype=fp8  
+            - --dtype=auto
            - --enable-prefix-caching
            - --tool-call-parser=qwen3_xml
            - --reasoning-parser=qwen3