Skip to content

fix lora target all + csgmv backend #14796

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
Fridge003 merged 3 commits into from
Dec 10, 2025
Merged

fix lora target all + csgmv backend #14796

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
cf293d5
more
b8zhong Dec 10, 2025
86843a2
more
b8zhong Dec 10, 2025
74b65c7
Merge branch 'main' into brayden/fix-lora-ci
b8zhong Dec 10, 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: 11 additions & 0 deletions python/sglang/srt/server_args.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4297,6 +4297,17 @@ def check_lora_server_args(self):
), "If 'all' is specified in --lora-target-modules, it should be the only module specified."
self.lora_target_modules = set(SUPPORTED_LORA_TARGET_MODULES)

# When using the chunked SGMV backend, skip embedding / lm_head layers for now,
# since it does not support these yet (TODO: implement embedding / lm_head support)
if self.lora_backend == "csgmv":
logger.warning(
"LoRA backend 'csgmv' does not yet support embedding or lm_head layers; "
"dropping 'embed_tokens' and 'lm_head' from --lora-target-modules=all. "
"To apply LoRA to these, use --lora-backend triton."
)
self.lora_target_modules.discard("embed_tokens")
self.lora_target_modules.discard("lm_head")

# Ensure sufficient information is provided for LoRA initialization.
assert self.lora_paths or (
self.max_lora_rank and self.lora_target_modules
Expand Down
Loading