Skip to content

[Bugfix] EPLB verify #4560

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

Closed
wants to merge 3 commits into from
Closed

[Bugfix] EPLB verify #4560

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
53994c3
[Bugfix] EPLB verfiy
shenchuxiaofugui Nov 29, 2025
433b924
cc
shenchuxiaofugui Nov 29, 2025
2af1086
cc
shenchuxiaofugui Nov 29, 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
10 changes: 6 additions & 4 deletions vllm_ascend/ops/common_fused_moe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -172,6 +172,7 @@ def __init__(self, *args, **kwargs):
self.expert_map_path = ascend_config.expert_map_path
self.global_redundant_expert_num = ascend_config.init_redundancy_expert
self.global_num_experts = num_experts + self.global_redundant_expert_num
init_eplb_enable = False
if self.custom_routing_function is None and self.e_score_correction_bias is not None:
vllm_config = get_current_vllm_config()
self.e_score_correction_bias.data = self.e_score_correction_bias.data.to(
Expand All @@ -191,6 +192,7 @@ def __init__(self, *args, **kwargs):
self.moe_instance_id, self.ep_rank))
self.log2phy = self.expert_load_balancer.get_rank_log2phy_map(
self.moe_instance_id, self.ep_rank).npu()
init_eplb_enable = True
except Exception as e:
logger.warning(
f"Init expert map of mtp/eagle when using sample.{e}")
Expand Down Expand Up @@ -236,10 +238,10 @@ def __init__(self, *args, **kwargs):
self.moe_load = torch.zeros(local_num_experts,
dtype=torch.int64).npu()

eplb_enable = self.dynamic_eplb or (self.expert_map_path is not None)
if eplb_enable and (not hasattr(self.quant_method, "quant_method") or
not isinstance(self.quant_method.quant_method,
AscendW8A8DynamicFusedMoEMethod)):
if init_eplb_enable and (
not hasattr(self.quant_method, "quant_method")
or not isinstance(self.quant_method.quant_method,
AscendW8A8DynamicFusedMoEMethod)):
raise ValueError("Eplb supports only w8a8_dynamic quantization.")

self.moe_config.num_experts = self.global_num_experts
Expand Down
Loading
Loading