[misc] Use --cuda-graph-max-bs-decode in tests, examples, and docs (#29591)

2026-06-28 18:38:28 -07:00
parent 3217410cf6
commit 909123ddb8
140 changed files with 304 additions and 290 deletions
@@ -25,7 +25,7 @@ class TestFlashAttention3LocalAttn(CustomTestCase):
    @classmethod
    def get_server_args(cls):
        return [
-            "--cuda-graph-max-bs",
+            "--cuda-graph-max-bs-decode",
            "2",
            "--attention-backend",
            "fa3",