[Moe Refactor] Make Inplace Flag for FusedMoEModularKernel part of the constructor (#33375)

Signed-off-by: Bill Nell <bnell@redhat.com> Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
2026-02-05 13:07:18 -05:00
parent 1ee95841bd
commit a57c8228ff
37 changed files with 132 additions and 109 deletions
--- a/tests/kernels/moe/test_flashinfer.py
+++ b/tests/kernels/moe/test_flashinfer.py
@@ -301,6 +301,7 @@ def test_flashinfer_cutlass_moe_fp8_no_graph(
                moe_config=moe_config,
                quant_config=quant_config,
            ),
+            inplace=False,
        )

        flashinfer_cutlass_output = kernel(
@@ -309,7 +310,6 @@ def test_flashinfer_cutlass_moe_fp8_no_graph(
            td.layer.w2_weight,
            topk_weights,
            topk_ids,
-            inplace=False,
            activation=activation,
            global_num_experts=e,
            expert_map=None,