[new feat] ascend backend support fia fusion kernel (#8328)

Co-authored-by: Even Zhou <even.y.zhou@outlook.com>
2025-08-26 14:13:08 +08:00
parent e2e378caba
commit f92b729d52
9 changed files with 509 additions and 128 deletions
--- a/test/srt/ascend/test_ascend_mla_w8a8int8.py
+++ b/test/srt/ascend/test_ascend_mla_w8a8int8.py
@@ -40,6 +40,7 @@ class TestAscendMlaW8A8Int8(CustomTestCase):
            "w8a8_int8",
            "--tp-size",
            4,
+            "--disable-radix-cache",
        ]

    def test_a_gsm8k(self):