Skip to content

[https://nvbugs/5488576][fix] Propagate disable_finalize_fusion config flag in WIDEEP MoE backend (cherry-pick #8141) #8566

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Oct 22, 2025
Merged

[https://nvbugs/5488576][fix] Propagate disable_finalize_fusion config flag in WIDEEP MoE backend (cherry-pick #8141) #8566

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 3 additions & 1 deletion tensorrt_llm/_torch/modules/fused_moe/fused_moe_wide_ep.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -222,6 +222,8 @@ def __init__(
f"Not available alltoall method type: {self.alltoall_method_type!r}"
)

self.use_fused_finalize = not model_config.moe_disable_finalize_fusion

self._weights_created = False
if not model_config.skip_create_weights_in_init:
self.create_weights()
Expand Down Expand Up @@ -689,7 +691,7 @@ def forward_chunk(
input_sf=x_sf,
swizzled_input_sf=False,
min_latency_mode=False,
use_fused_finalize=True,
use_fused_finalize=self.use_fused_finalize,
tuner_num_tokens=tuner_num_tokens,
tuner_top_k=tuner_top_k,
)
Expand Down
Loading