Skip to content

[Frontend] Reduce chat template warmup logging levels #37062

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
njhill merged 2 commits into from
Mar 14, 2026
+7 −8
Merged

[Frontend] Reduce chat template warmup logging levels #37062

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5bd0082
[Frontend] Reduce chat template warmup logging levels
njhill Mar 14, 2026
01a38ca
couple more
njhill Mar 14, 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
15 changes: 7 additions & 8 deletions vllm/renderers/base.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -179,17 +179,17 @@ def warmup(self, chat_params: ChatParams) -> None:
from vllm.entrypoints.chat_utils import ChatTemplateResolutionError

try:
logger.info("Warming up chat template processing...")
logger.debug("Warming up chat template processing...")
start_time = time.perf_counter()

self.render_chat([[{"role": "user", "content": "warmup"}]], chat_params)

elapsed = time.perf_counter() - start_time
logger.info("Chat template warmup completed in %.3fs", elapsed)
logger.debug("Chat template warmup completed in %.3fs", elapsed)
except ChatTemplateResolutionError:
logger.info("This model does not support chat template.")
logger.debug("This model does not support chat template.")
except Exception:
logger.exception("Chat template warmup failed")
logger.warning("Chat template warmup failed", exc_info=True)

if self.mm_processor:
from vllm.multimodal.processing import TimingContext
Expand All @@ -200,7 +200,7 @@ def warmup(self, chat_params: ChatParams) -> None:
mm_limits = processor.info.allowed_mm_limits

try:
logger.info("Warming up multi-modal processing...")
logger.debug("Warming up multi-modal processing...")
start_time = time.perf_counter()

processor_inputs = processor.dummy_inputs.get_dummy_processor_inputs(
Expand All @@ -209,14 +209,13 @@ def warmup(self, chat_params: ChatParams) -> None:
mm_options=mm_config.limit_per_prompt,
)
_ = processor.apply(
processor_inputs,
timing_ctx=TimingContext(enabled=False),
processor_inputs, timing_ctx=TimingContext(enabled=False)
)

elapsed = time.perf_counter() - start_time
logger.info("Multi-modal warmup completed in %.3fs", elapsed)
except Exception:
logger.exception("Multi-modal warmup failed")
logger.warning("Multi-modal warmup failed")
finally:
self.clear_mm_cache()

Expand Down
Loading