[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>
This commit is contained in:
bnellnm
2026-02-05 13:07:18 -05:00
committed by GitHub
parent 1ee95841bd
commit a57c8228ff
37 changed files with 132 additions and 109 deletions

View File

@@ -179,7 +179,11 @@ def make_modular_kernel(
quant_config=quant_config,
)
mk = FusedMoEModularKernel(prepare_finalize=a2a, fused_experts=fused_experts)
mk = FusedMoEModularKernel(
prepare_finalize=a2a,
fused_experts=fused_experts,
inplace=False,
)
return mk
@@ -256,7 +260,6 @@ def deep_ep_moe_impl(
w2=w2,
topk_weights=topk_weights_chunk,
topk_ids=topk_chunk,
inplace=False,
activation="silu",
global_num_experts=num_experts,
expert_map=build_expert_map(),