Skip to content

fix kimi k2.5's moe gemm config init #18064

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
Kangyan-Zhou merged 4 commits into from
Feb 5, 2026
Merged

fix kimi k2.5's moe gemm config init #18064

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
5626633
fix kimi k2.5's moe gemm config init
cicirori Feb 1, 2026
83d79e4
Merge branch 'main' into fix_kimi_k25_moe_init
cicirori Feb 1, 2026
b6c3c58
fix kimi k2.5's moe gemm config init
cicirori Feb 1, 2026
27abe08
Merge branch 'main' into fix_kimi_k25_moe_init
cicirori Feb 4, 2026
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
7 changes: 6 additions & 1 deletion python/sglang/srt/managers/scheduler.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -485,7 +485,12 @@ def init_tokenizer(self):
)[0]

def init_moe_gemm_config(self):
if hasattr(self.model_config.hf_config, "num_experts_per_tok"):
# For the MM models, check the text_config for MoE settings
config_to_check = getattr(
self.model_config.hf_config, "text_config", self.model_config.hf_config
)

if hasattr(config_to_check, "num_experts_per_tok"):
initialize_moe_config(self.server_args)

# Initialize GEMM-related configuration for FP8 and FP4 backends.
Expand Down
Loading