Update main.py

main.py

@@ -120,27 +120,55 @@ async def fetch_models():
 @app.post("/chat/completions")
 @app.post("/v1/chat/completions")
 async def get_completion(payload: Payload, request: Request):
-    # Use the correct endpoint depending on the model type (no authentication now 😉)
-    endpoint = secret_api_endpoint_2 if payload.model in alternate_models else secret_api_endpoint
-
-    # Use the payload directly as it includes stream and other user data
+    model_to_use = "gpt-4o-mini" if payload.model == "gpt-4o" else payload.model
     payload_dict = payload.dict()
-    print(payload_dict)  # coz I m curious af heheheh :)
-    # data is kept to me only don't worry :)
+    payload_dict["model"] = model_to_use
+
+    # Select the appropriate endpoint
+    endpoint = secret_api_endpoint_2 if model_to_use in alternate_models else secret_api_endpoint
+
+    print(payload_dict)
 
     async def stream_generator(payload_dict):
         async with httpx.AsyncClient() as client:
             try:
                 async with client.stream("POST", f"{endpoint}/v1/chat/completions", json=payload_dict, timeout=10) as response:
-                    response.raise_for_status()
+                    if response.status_code == 422:
+                        # Handle unprocessable entity errors
+                        raise HTTPException(status_code=422, detail="Unprocessable entity. Check your payload.")
+                    elif response.status_code == 400:
+                        # Handle bad request errors
+                        raise HTTPException(status_code=400, detail="Bad request. Verify input data.")
+                    elif response.status_code == 403:
+                        # Handle forbidden access
+                        raise HTTPException(status_code=403, detail="Forbidden. You do not have access to this resource.")
+                    elif response.status_code == 404:
+                        # Handle not found errors
+                        raise HTTPException(status_code=404, detail="The requested resource was not found.")
+                    elif response.status_code >= 500:
+                        # Handle server errors
+                        raise HTTPException(status_code=500, detail="Server error. Try again later.")
+                    
+                    response.raise_for_status()  # Raise HTTPStatusError for non-200 responses not explicitly handled
+                    
+                    # Stream response to the client
                     async for line in response.aiter_lines():
                         if line:
                             yield f"{line}\n"
             except httpx.HTTPStatusError as status_err:
-                raise HTTPException(status_code=status_err.response.status_code, detail=f"HTTP error: {status_err}")
+                # Catch specific HTTP errors
+                raise HTTPException(
+                    status_code=status_err.response.status_code,
+                    detail=f"HTTP error: {status_err.response.text}"
+                )
+            except httpx.TimeoutException:
+                # Handle timeout errors
+                raise HTTPException(status_code=504, detail="Request timed out. Please try again later.")
             except httpx.RequestError as req_err:
-                raise HTTPException(status_code=500, detail=f"Streaming failed: {req_err}")
+                # Handle generic request errors
+                raise HTTPException(status_code=500, detail=f"Request failed: {req_err}")
             except Exception as e:
+                # Catch any unexpected exceptions
                 raise HTTPException(status_code=500, detail=f"An unexpected error occurred: {e}")
 
     return StreamingResponse(stream_generator(payload_dict), media_type="application/json")