[Model] Use explicit types in get_generation_prompt (#33551)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

vllm/model_executor/models/qwen3_asr.py

@@ -23,7 +23,7 @@
 """Inference-only Qwen3-ASR model."""
 
 from collections.abc import Iterable, Mapping, Sequence
-from typing import Any, Literal, cast
+from typing import Any, Literal
 
 import numpy as np
 import torch
@@ -33,7 +33,7 @@ from transformers.models.whisper import WhisperFeatureExtractor
 
 from vllm.config import ModelConfig, SpeechToTextConfig, VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
-from vllm.inputs.data import PromptType
+from vllm.inputs.data import PromptType, TokensPrompt
 from vllm.logger import init_logger
 from vllm.model_executor.models.interfaces import (
     MultiModalEmbeddings,
@@ -561,11 +561,11 @@ class Qwen3ASRForConditionalGeneration(
             )
 
         prompt_token_ids = tokenizer.encode(prompt)
-        prompt_dict = {
-            "prompt_token_ids": prompt_token_ids,
-            "multi_modal_data": {"audio": audio},
-        }
-        return cast(PromptType, prompt_dict)
+
+        return TokensPrompt(
+            prompt_token_ids=prompt_token_ids,
+            multi_modal_data={"audio": audio},
+        )
 
     @classmethod
     def post_process_output(cls, text: str) -> str: