Remove chat_service.py abstraction layer

- Remove unnecessary chat_service.py pass-through layer
- Update router to call transformers_provider directly
- Update tests to mock provider functions instead of service layer
- Simplify architecture: Router → Provider → Model

app/routers/openai_api.py

@@ -6,8 +6,7 @@ from fastapi.responses import StreamingResponse, JSONResponse
 
 from app.config import settings
 from app.models.openai import ChatCompletionRequest
-from app.services import chat_service
-from app.providers.transformers_provider import initialize_model
+from app.providers.transformers_provider import initialize_model, chat, list_models
 
 logger = logging.getLogger(__name__)
 
@@ -15,9 +14,9 @@ router = APIRouter()
 
 
 @router.get("/models")
-async def list_models():
+async def list_models_endpoint():
     """List available models (OpenAI-compatible endpoint)"""
-    return await chat_service.list_models()
+    return await list_models()
 
 
 @router.post("/models/reload")
@@ -115,12 +114,12 @@ async def chat_completions(body: ChatCompletionRequest):
         logger.info(f"Chat completion request: model={payload['model']}, messages={len(payload['messages'])}, stream={payload['stream']}")
 
         if body.stream:
-            stream = await chat_service.chat(payload, stream=True)
+            stream = await chat(payload, stream=True)
             # stream is already an AsyncIterator[str] with SSE-formatted chunks
             return StreamingResponse(stream, media_type="text/event-stream")
 
         # Non-streaming response
-        data = await chat_service.chat(payload, stream=False)
+        data = await chat(payload, stream=False)
         return JSONResponse(content=data)
         
     except ValueError as e:

app/services/chat_service.py

@@ -1,33 +0,0 @@
-"""Chat service layer providing abstraction over the provider."""
-from typing import Any, Dict, Union, AsyncIterator
-
-from app.providers import transformers_provider as provider
-
-
-async def list_models() -> Dict[str, Any]:
-    """
-    List available models.
-    
-    Returns:
-        Dictionary containing model list in OpenAI-compatible format
-    """
-    return await provider.list_models()
-
-
-async def chat(
-    payload: Dict[str, Any], 
-    stream: bool = False
-) -> Union[Dict[str, Any], AsyncIterator[str]]:
-    """
-    Process chat completion request.
-    
-    Args:
-        payload: Request payload containing messages and generation parameters
-        stream: Whether to stream the response
-    
-    Returns:
-        Response dictionary or async iterator for streaming
-    """
-    return await provider.chat(payload, stream=stream)
-
-

tests/test_openai_routes.py

@@ -10,9 +10,9 @@ def test_models(monkeypatch):
     async def fake_list_models():
         return {"data": [{"id": "DragonLLM/LLM-Pro-Finance-Small"}]}
 
-    from app.services import chat_service
+    from app.providers import transformers_provider
 
-    monkeypatch.setattr(chat_service, "list_models", fake_list_models)
+    monkeypatch.setattr(transformers_provider, "list_models", fake_list_models)
 
     r = client.get("/v1/models")
     assert r.status_code == 200
@@ -37,9 +37,9 @@ def test_chat_completions(monkeypatch):
             ],
         }
 
-    from app.services import chat_service
+    from app.providers import transformers_provider
 
-    monkeypatch.setattr(chat_service, "chat", fake_chat)
+    monkeypatch.setattr(transformers_provider, "chat", fake_chat)
 
     r = client.post(
         "/v1/chat/completions",