[CI] test chunked prefill more (#5798)

This commit is contained in:
Lianmin Zheng
2025-04-28 10:57:17 -07:00
committed by GitHub
parent d73ddeb196
commit 849c83a0c0
15 changed files with 212 additions and 97 deletions

View File

@@ -26,6 +26,8 @@ class TestMLA(CustomTestCase):
"--enable-torch-compile",
"--cuda-graph-max-bs",
"2",
"--chunked-prefill-size",
"256",
],
)