[Frontend]: Add apply_chat_template method and update generate method in LLM class

vllm/entrypoints/llm.py

@@ -1,5 +1,6 @@
 from contextlib import contextmanager
-from typing import ClassVar, List, Optional, Sequence, Union, cast, overload
+from typing import (ClassVar, Dict, List, Optional, Sequence, Union, cast,
+                    overload)
 
 from tqdm import tqdm
 from transformers import PreTrainedTokenizer, PreTrainedTokenizerFast
@@ -173,6 +174,35 @@ def set_tokenizer(
             self.llm_engine.tokenizer.tokenizer = get_cached_tokenizer(
                 tokenizer)
 
+    def apply_chat_template(
+        self,
+        messages_list: Union[List[List[Dict[str, str]]], List[Dict[str, str]]],
+        add_generation_prompt: bool = True,
+        tokenize: bool = False
+    ) -> Union[Union[List[List[str]], List[str]], Union[List[List[int]],
+                                                        List[int]]]:
+        """Applies a chat template to the given messages.
+
+        Args:
+            messages_list: A list of list of dicts of list of dicts messages to
+            be processed. add_generation_prompt: Whether to add a generation
+            prompt at the end.
+            tokenize: Whether to return token IDs (True) or
+            raw text (False).
+
+        Returns:
+            A list of token IDs if tokenize is True, otherwise a
+            list of text strings.
+        """
+        tokenizer = self.get_tokenizer()
+        ids_or_text = [
+            tokenizer.apply_chat_template(
+                messages,
+                add_generation_prompt=add_generation_prompt,
+                tokenize=tokenize) for messages in messages_list
+        ]
+        return ids_or_text
+
     @overload  # LEGACY: single (prompt + optional token ids)
     def generate(
         self,
@@ -259,6 +289,8 @@ def generate(
         sampling_params: Optional[Union[SamplingParams,
                                         Sequence[SamplingParams]]] = None,
         prompt_token_ids: Optional[Union[List[int], List[List[int]]]] = None,
+        messages_list: Optional[Union[List[List[Dict[str, str]]],
+                                      List[Dict[str, str]]]] = None,
         use_tqdm: bool = True,
         lora_request: Optional[Union[List[LoRARequest], LoRARequest]] = None,
         prompt_adapter_request: Optional[PromptAdapterRequest] = None,
@@ -295,6 +327,12 @@ def generate(
                 "LLM.generate() is only supported for generation models "
                 "(XForCausalLM).")
 
+        # Use apply_chat_template if prompt_token_ids is not provided
+        if prompt_token_ids is None and messages_list is not None:
+            prompt_token_ids = cast(
+                Union[List[int], List[List[int]]],
+                self.apply_chat_template(messages_list, tokenize=True))
+
         if prompt_token_ids is not None:
             inputs = self._convert_v1_inputs(
                 prompts=cast(Optional[Union[str, List[str]]], prompts),