Add env var to control custom Triton kernel cache and set CSGMV as default backend. (#12176)

2025-10-27 17:49:32 -07:00
parent 88596739a4
commit ce832d7034
13 changed files with 24 additions and 25 deletions
@@ -62,7 +62,6 @@ class TestLoRACudaGraph(CustomTestCase):
                    model_case,
                    torch_dtype,
                    max_new_tokens=32,
-                    backend="triton",
                    disable_cuda_graph=True,
                    test_tag="without_cuda_graph",
                )
@@ -77,7 +76,6 @@ class TestLoRACudaGraph(CustomTestCase):
                    model_case,
                    torch_dtype,
                    max_new_tokens=32,
-                    backend="triton",
                    disable_cuda_graph=False,
                    test_tag="cuda_graph_padding",
                )