Update llm.py based on Code Review

Author: NekoMimiUnagi

vllm/entrypoints/llm.py

@@ -604,18 +604,13 @@ def create_tokens_prompt_from_beam(
                     **mm_kwargs,
                 ), )
 
-        token_iter = range(max_tokens)
-        if use_tqdm:
-            token_iter = tqdm(range(max_tokens, desc="Beam Searching"))
-            warnings.warn(
-                "The progress bar shows the upper bound on token steps and "
-                "may finish early due to stopping conditions. It does not "
-                "reflect instance-level progress.",
-                stacklevel=2,
-                unit="token",
-                unit_scale=False,
-            )
-          
+        token_iter = tqdm(range(max_tokens), desc="Beam search", unit="token", unit_scale=False)
+        warnings.warn(
+            "The progress bar shows the upper bound on token steps and "
+            "may finish early due to stopping conditions. It does not "
+            "reflect instance-level progress.",
+            stacklevel=2,
+        )          
         for _ in token_iter:
             all_beams: list[BeamSearchSequence] = list(
                 sum((instance.beams for instance in instances), []))