Update app.py

app.py

@@ -67,6 +67,7 @@ class ModelManager:
                 self.models[model_config['name']] = Llama.from_pretrained(repo_id=model_config['repo_id'], filename=model_config['filename'], use_auth_token=HUGGINGFACE_TOKEN)
             except Exception as e:
                 print(f"Error loading model {model_config['name']}: {e}")
+                pass  # Add pass to handle exceptions during model loading
 
     def load_all_models(self):
         with ThreadPoolExecutor() as executor:
@@ -109,9 +110,11 @@ def generate_model_response(model, inputs):
                 return remove_duplicates(response['choices'][0]['text'])
             except Exception as e2:
                 print(f"Error generating model response (after retry): {e2}")
+                pass  # Add pass to handle exceptions during retry
                 return ""
         else:
             print(f"Error generating model response: {e}")
+            pass  # Add pass to handle other exceptions
             return ""
 
 def remove_repetitive_responses(responses):
@@ -139,10 +142,14 @@ async def process_message(message):
 @app.post("/generate_multimodel")
 async def api_generate_multimodel(request: Request):
     while True:
-        data = await request.json()
-        message = data["message"]
-        formatted_response = await process_message(message)
-        return JSONResponse({"response": formatted_response})
+        try: 
+            data = await request.json()
+            message = data["message"]
+            formatted_response = await process_message(message)
+            return JSONResponse({"response": formatted_response})
+        except Exception as e:
+            print(f"Error in API request handling: {e}")
+            pass  # Add pass to handle exceptions in API request handling
         time.sleep(300)
 
 iface = gr.Interface(