[Feature] Batch invariant: Enable TRITON_MLA without prefix-caching (#29125)

Signed-off-by: yewentao256 <zhyanwentao@126.com>
2025-12-08 19:31:57 -05:00
parent 9d6235ca9a
commit d9417096d1
5 changed files with 43 additions and 7 deletions
--- a/tests/v1/determinism/utils.py
+++ b/tests/v1/determinism/utils.py
@@ -17,6 +17,7 @@ skip_unsupported = pytest.mark.skipif(

 BACKENDS: list[str] = [
    "FLASH_ATTN",
+    "TRITON_MLA",
 ]

 if has_flashinfer():