added openAI schema based endpoint and response v5

main/api.py

@@ -19,10 +19,11 @@ class InferenceApi(LitAPI):
         self.config = config
         self.llm_config = config.get('llm_server', {})
 
-    async def setup(self, device: Optional[str] = None):
-        """Setup method required by LitAPI"""
+    def setup(self, device: Optional[str] = None):
+        """Synchronous setup method required by LitAPI"""
         self._device = device
         self.logger.info(f"Inference API setup completed on device: {device}")
+        return self  # It's common for setup methods to return self for chaining
 
     async def _get_client(self):
         """Get or create HTTP client as needed"""