Skip to content

[Bugfix] dynamic eplb does't use fused_alltoall #4919

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
MengqingCao merged 2 commits into from
Dec 16, 2025
+7 −4
Merged

[Bugfix] dynamic eplb does't use fused_alltoall #4919

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
9c360a3
[Bugfix] dynamic eplb does't use fused_alltoall
shenchuxiaofugui Dec 11, 2025
5e76461
Merge branch 'main' into ffn_fused
MengqingCao Dec 16, 2025
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
11 changes: 7 additions & 4 deletions vllm_ascend/worker/model_runner_v1.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1434,10 +1434,13 @@ def _select_moe_comm_method(self,
moe_comm_type = MoECommType.ALLGATHER

elif soc_version in {AscendDeviceType._910_93}:
moe_comm_type = (
MoECommType.MC2 if num_tokens <= mc2_tokens_capacity else
MoECommType.FUSED_ALLTOALL if quant_type == "w8a8_dynamic"
and get_ep_group().world_size <= 16 else MoECommType.ALLTOALL)
# TODO: drop the EP-size guard when dispatch_ffn_combine supports larger EP sizes
fused_all2all_enable = quant_type == "w8a8_dynamic" and get_ep_group(
).world_size <= 16 and (not self.dynamic_eplb)
moe_comm_type = (MoECommType.MC2
if num_tokens <= mc2_tokens_capacity else
MoECommType.FUSED_ALLTOALL
if fused_all2all_enable else MoECommType.ALLTOALL)
else:
raise ValueError(f"Unsupported soc_version: {soc_version}")

Expand Down
Loading