From 2010f04c17e76c7d1f70f6e1c9d3857a93036114 Mon Sep 17 00:00:00 2001 From: Jee Jee Li Date: Mon, 17 Feb 2025 11:26:24 +0800 Subject: [PATCH] [V1][Misc] Avoid unnecessary log output (#13289) --- vllm/v1/worker/gpu_model_runner.py | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/vllm/v1/worker/gpu_model_runner.py b/vllm/v1/worker/gpu_model_runner.py index d3995b619d31..f1212c3554b6 100644 --- a/vllm/v1/worker/gpu_model_runner.py +++ b/vllm/v1/worker/gpu_model_runner.py @@ -96,11 +96,13 @@ class GPUModelRunner(LoRAModelRunnerMixin): self.mm_registry = MULTIMODAL_REGISTRY self.uses_mrope = model_config.uses_mrope - # NOTE: Initialized client is only used for processing dummy - # multimodal data into multimodal kwargs for GPU memory profiling. - # Only applicable to multimodal models with legacy input mapper. - self.mm_input_mapper_profiling = MMInputCacheClient(self.model_config) - self.mm_input_mapper_profiling.use_cache = False + if self.is_multimodal_model: + # NOTE: Initialized client is only used for processing dummy + # multimodal data into multimodal kwargs for GPU memory profiling. + # Only applicable to multimodal models with legacy input mapper. + self.mm_input_mapper_profiling = MMInputCacheClient( + self.model_config) + self.mm_input_mapper_profiling.use_cache = False encoder_compute_budget, encoder_cache_size = compute_encoder_budget( model_config=model_config,