Log not important

Signed-off-by: Diego-Castan <[email protected]>

Author: diegocastanibm

vllm/v1/worker/gpu_worker.py

@@ -383,6 +383,7 @@ def compile_cuda_graph(input_size: int):
         # but users still want to compile for better performance,
         # e.g. for the max-num-batched token size in chunked prefill.
         warmup_sizes = self.vllm_config.compilation_config.compile_sizes.copy()
+        logger.info("Warm up sizes %s", str(warmup_sizes))
         if not self.model_config.enforce_eager:
             warmup_sizes = [
                 x for x in warmup_sizes if x not in