Update app.py

app.py

@@ -56,7 +56,13 @@ class ModelManager:
     def load_model(self, model_config):
         if model_config['name'] not in self.models:
             try:
-                self.models[model_config['name']] = Llama.from_pretrained(repo_id=model_config['repo_id'], filename=model_config['filename'], use_auth_token=HUGGINGFACE_TOKEN)
+                print(f"Loading model {model_config['name']}...")
+                self.models[model_config['name']] = Llama.from_pretrained(
+                    repo_id=model_config['repo_id'], 
+                    filename=model_config['filename'], 
+                    use_auth_token=HUGGINGFACE_TOKEN
+                )
+                print(f"Model {model_config['name']} loaded successfully.")
             except Exception as e:
                 print(f"Error loading model {model_config['name']}: {e}")
 
@@ -91,11 +97,13 @@ def remove_duplicates(text):
 @spaces.GPU(queue=False, allow_gpu_memory=True, timeout=0, duration=0)
 def generate_model_response(model, inputs):
     try:
+        print(f"Generating response for model: {model}")
         response = model(inputs)
+        print(f"Response from {model}: {response}")
         return remove_duplicates(response['choices'][0]['text'])
     except Exception as e:
-        print(f"Error generating model response: {e}")
-        return ""
+        print(f"Error generating model response from {model}: {e}")
+        return "Error generating response."
 
 def remove_repetitive_responses(responses):
     unique_responses = {}
@@ -111,7 +119,15 @@ async def process_message(message):
             executor.submit(generate_model_response, model, inputs)
             for model in global_data['models'].values()
         ]
-        responses = [{'model': model_name, 'response': future.result()} for model_name, future in zip(global_data['models'].keys(), as_completed(futures))]
+        responses = []
+        for future in as_completed(futures):
+            try:
+                response = future.result()
+                responses.append(response)
+            except Exception as e:
+                print(f"Error with model: {e}")
+                responses.append("Error generating response.")  # O un mensaje predeterminado de error
+
     unique_responses = remove_repetitive_responses(responses)
     formatted_response = ""
     for model, response in unique_responses.items():