diff --git a/vllm/v1/worker/cpu_model_runner.py b/vllm/v1/worker/cpu_model_runner.py index e6fcac8fc5..e5800da98b 100644 --- a/vllm/v1/worker/cpu_model_runner.py +++ b/vllm/v1/worker/cpu_model_runner.py @@ -8,7 +8,10 @@ import torch import torch.nn as nn import vllm.utils.cpu_triton_utils as cpu_tl -from vllm.config import VllmConfig +from vllm.config import ( + CompilationMode, + VllmConfig, +) from vllm.logger import init_logger from vllm.model_executor.model_loader import get_model from vllm.tracing import instrument @@ -142,6 +145,8 @@ class CPUModelRunner(GPUModelRunner): @instrument(span_name="Warmup (CPU)") def warming_up_model(self) -> None: + if self.vllm_config.compilation_config.mode == CompilationMode.NONE: + return logger.info("Warming up model for the compilation...") # Only generate graph for the generic shape with _set_global_compilation_settings(self.vllm_config):