data_parallel_size in in VllmserveCliArgs (#3074)

* data_parallel_size in in VllmserveCliArgs * moved to 43
2025-08-18 18:14:37 +05:30
parent 0eef385b1a
commit c10eb811fa
1 changed files with 6 additions and 0 deletions
--- a/src/axolotl/cli/args.py
+++ b/src/axolotl/cli/args.py
@@ -40,6 +40,12 @@ class VllmServeCliArgs:
        default=None,
        metadata={"help": "Number of tensor parallel workers to use."},
    )
+    data_parallel_size: Optional[int] = field(
+        default=None,
+        metadata={
+            "help": "Number of data parallel workers to use for vLLM serving. This controls how many model replicas are used for parallel inference."
+        },
+    )
    host: Optional[str] = field(
        default=None,  # nosec B104
        metadata={"help": "Host address to run the server on."},