[Bugfix][Refactor] Unify model management in frontend (#11660)

Signed-off-by: Joe Runde <Joseph.Runde@ibm.com>

vllm/entrypoints/openai/serving_chat.py

@@ -21,10 +21,8 @@ from vllm.entrypoints.openai.protocol import (
     ChatCompletionStreamResponse, ChatMessage, DeltaFunctionCall, DeltaMessage,
     DeltaToolCall, ErrorResponse, FunctionCall, PromptTokenUsageInfo,
     RequestResponseMetadata, ToolCall, UsageInfo)
-from vllm.entrypoints.openai.serving_engine import (BaseModelPath,
-                                                    LoRAModulePath,
-                                                    OpenAIServing,
-                                                    PromptAdapterPath)
+from vllm.entrypoints.openai.serving_engine import OpenAIServing
+from vllm.entrypoints.openai.serving_models import OpenAIServingModels
 from vllm.entrypoints.openai.tool_parsers import ToolParser, ToolParserManager
 from vllm.logger import init_logger
 from vllm.outputs import CompletionOutput, RequestOutput
@@ -42,11 +40,9 @@ class OpenAIServingChat(OpenAIServing):
         self,
         engine_client: EngineClient,
         model_config: ModelConfig,
-        base_model_paths: List[BaseModelPath],
+        models: OpenAIServingModels,
         response_role: str,
         *,
-        lora_modules: Optional[List[LoRAModulePath]],
-        prompt_adapters: Optional[List[PromptAdapterPath]],
         request_logger: Optional[RequestLogger],
         chat_template: Optional[str],
         chat_template_content_format: ChatTemplateContentFormatOption,
@@ -57,9 +53,7 @@ class OpenAIServingChat(OpenAIServing):
     ) -> None:
         super().__init__(engine_client=engine_client,
                          model_config=model_config,
-                         base_model_paths=base_model_paths,
-                         lora_modules=lora_modules,
-                         prompt_adapters=prompt_adapters,
+                         models=models,
                          request_logger=request_logger,
                          return_tokens_as_token_ids=return_tokens_as_token_ids)
 
@@ -126,7 +120,7 @@ class OpenAIServingChat(OpenAIServing):
                 prompt_adapter_request,
             ) = self._maybe_get_adapters(request)
 
-            model_name = self._get_model_name(lora_request)
+            model_name = self.models.model_name(lora_request)
 
             tokenizer = await self.engine_client.get_tokenizer(lora_request)