[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
@@ -69,7 +69,7 @@ class TestPCGWithNGRAM(PCGSpecBase, unittest.TestCase):
        "NGRAM",
        "--speculative-num-draft-tokens",
        "16",
-        "--cuda-graph-max-bs",
+        "--cuda-graph-max-bs-decode",
        "8",
        "--mem-fraction-static",
        "0.8",