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, 11:00 UTC
Remove noisy warnings from `SchedulerConfig` (#17995)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Harry Mellor committed
May 13, 2025 at 01:33 UTC
d67085c2c8a5de07fb531b0c0416d173f31cd9af
1 file changed
-6
vllm/config.py
-6
@@ -2018,15 +2018,9 @@ class SchedulerConfig:
2018
def __post_init__(self) -> None:
2019
if self.max_model_len is None:
2020
self.max_model_len = 8192
2021
- logger.warning_once(
2022
- "max_model_len was is not set. Defaulting to arbitrary value "
2023
- "of %d.", self.max_model_len)
2021
2022
if self.max_num_seqs is None:
2023
self.max_num_seqs = 128
2027
- logger.warning_once(
2028
- "max_num_seqs was is not set. Defaulting to arbitrary value "
2029
- "of %d.", self.max_num_seqs)
2024
2025
if self.max_num_batched_tokens is None:
2026
if self.enable_chunked_prefill: