From bfbcde26d5c86475d478b6b1abc9a1285d3753d0 Mon Sep 17 00:00:00 2001 From: Ning Xie Date: Thu, 1 Oct 2026 10:54:33 +0800 Subject: [PATCH] [Observability] add model initializing duration log (#52141) Signed-off-by: Andy Xie --- vllm/model_executor/model_loader/base_loader.py | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/vllm/model_executor/model_loader/base_loader.py b/vllm/model_executor/model_loader/base_loader.py index 7712efbfd1..2b7cb72fff 100644 --- a/vllm/model_executor/model_loader/base_loader.py +++ b/vllm/model_executor/model_loader/base_loader.py @@ -1,5 +1,6 @@ # SPDX-License-Identifier: Apache-2.0 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project +import time from abc import ABC, abstractmethod import torch @@ -70,11 +71,17 @@ class BaseModelLoader(ABC): target_device = torch.device(load_device) with set_default_torch_dtype(model_config.dtype): with target_device: + time_before_load = time.perf_counter() model = self.create_model( vllm_config=vllm_config, model_config=model_config, prefix=prefix, ) + time_after_load = time.perf_counter() + logger.info_once( + "Initializing model took %.6f seconds", + time_after_load - time_before_load, + ) logger.debug("Loading weights on %s ...", load_device) self.load_weights(model, model_config)