A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 12:15 UTC
Fix call to `logger.info_once` (#17416)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Harry Mellor committed
Apr 30, 2025 at 03:39 UTC
d1f569b1b9ce37aa77873b7f2bdb73612c4d9f23
1 file changed
+2
-2
vllm/model_executor/guided_decoding/xgrammar_decoding.py
+2
-2
@@ -191,9 +191,9 @@ class GrammarConfig:
191
192
if model_with_warn is not None and any_whitespace:
193
logger.info_once(
194
- "%s model detected, consider setting "
194
+ f"{model_with_warn} model detected, consider setting "
195
"`disable_any_whitespace` to prevent runaway generation "
196
- "of whitespaces.", model_with_warn)
196
+ "of whitespaces.")
197
# Validate the schema and raise ValueError here if it is invalid.
198
# This is to avoid exceptions in model execution, which will crash
199
# the engine worker process.