Fix AMX GQA extend attention (#25180)

Co-authored-by: Ma Mingfei <mingfei.ma@intel.com>
This commit is contained in:
miamiaoxyz
2026-05-18 09:30:02 +08:00
committed by GitHub
co-authored by Ma Mingfei
parent 2a357071ec
commit 5147de26e4
2 changed files with 15 additions and 2 deletions
+12
View File
@@ -217,6 +217,18 @@ class TestExtendAttention(CustomTestCase):
b_seq_len_extend=[5000],
)
def test_extend_attention_gqa_partial_extend_with_prefix(self):
self._test_extend_attention_once(
B=1,
N_CTX=256,
H_Q=16,
H_KV=4,
D=128,
DV=96,
b_seq_len_prefix=[97],
b_seq_len_extend=[37],
)
if __name__ == "__main__":
unittest.main()