[V0 Deprecation] Remove MQLLMEngine (#25019)

Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>

tests/entrypoints/openai/test_serving_chat.py

@@ -13,12 +13,12 @@ import pytest
 import pytest_asyncio
 
 from vllm.config.multimodal import MultiModalConfig
-from vllm.engine.multiprocessing.client import MQLLMEngineClient
 from vllm.entrypoints.openai.protocol import ChatCompletionRequest
 from vllm.entrypoints.openai.serving_chat import OpenAIServingChat
 from vllm.entrypoints.openai.serving_models import (BaseModelPath,
                                                     OpenAIServingModels)
 from vllm.transformers_utils.tokenizer import get_tokenizer
+from vllm.v1.engine.async_llm import AsyncLLM
 
 from ...utils import RemoteOpenAIServer
 
@@ -276,7 +276,7 @@ def test_async_serving_chat_init():
 
 @pytest.mark.asyncio
 async def test_serving_chat_returns_correct_model_name():
-    mock_engine = MagicMock(spec=MQLLMEngineClient)
+    mock_engine = MagicMock(spec=AsyncLLM)
     mock_engine.get_tokenizer.return_value = get_tokenizer(MODEL_NAME)
     mock_engine.errored = False
 
@@ -312,7 +312,7 @@ async def test_serving_chat_returns_correct_model_name():
 
 @pytest.mark.asyncio
 async def test_serving_chat_should_set_correct_max_tokens():
-    mock_engine = MagicMock(spec=MQLLMEngineClient)
+    mock_engine = MagicMock(spec=AsyncLLM)
     mock_engine.get_tokenizer.return_value = get_tokenizer(MODEL_NAME)
     mock_engine.errored = False
 
@@ -355,7 +355,7 @@ async def test_serving_chat_should_set_correct_max_tokens():
     }
 
     # Reinitialize the engine with new settings
-    mock_engine = MagicMock(spec=MQLLMEngineClient)
+    mock_engine = MagicMock(spec=AsyncLLM)
     mock_engine.get_tokenizer.return_value = get_tokenizer(MODEL_NAME)
     mock_engine.errored = False
 
@@ -410,7 +410,7 @@ async def test_serving_chat_should_set_correct_max_tokens():
     }
 
     # Reinitialize the engine with new settings
-    mock_engine = MagicMock(spec=MQLLMEngineClient)
+    mock_engine = MagicMock(spec=AsyncLLM)
     mock_engine.get_tokenizer.return_value = get_tokenizer(MODEL_NAME)
     mock_engine.errored = False
 
@@ -467,7 +467,7 @@ async def test_serving_chat_could_load_correct_generation_config():
         "repetition_penalty": 1.05
     }
 
-    mock_engine = MagicMock(spec=MQLLMEngineClient)
+    mock_engine = MagicMock(spec=AsyncLLM)
     mock_engine.get_tokenizer.return_value = get_tokenizer(MODEL_NAME)
     mock_engine.errored = False
 
@@ -523,7 +523,7 @@ async def test_serving_chat_did_set_correct_cache_salt(model_type):
     mock_model_config = MockModelConfig()
     mock_model_config.hf_config.model_type = model_type
 
-    mock_engine = MagicMock(spec=MQLLMEngineClient)
+    mock_engine = MagicMock(spec=AsyncLLM)
     mock_engine.get_tokenizer.return_value = get_tokenizer(MODEL_NAME)
     mock_engine.errored = False