Update app.py

app.py

@@ -53,9 +53,12 @@ class ModelManager:
         self.models = []
         self.loaded = False
 
-    @spaces.GPU(duration=0)
     def load_model(self, model_config):
-        return {"model": Llama.from_pretrained(repo_id=model_config['repo_id'], filename=model_config['filename']), "name": model_config['name']}
+        try:
+            return {"model": Llama.from_pretrained(repo_id=model_config['repo_id'], filename=model_config['filename']), "name": model_config['name']}
+        except Exception as e:
+            print(f"Error loading model {model_config['name']}: {e}")
+            return None
     
     def load_all_models(self):
         if self.loaded:
@@ -65,11 +68,9 @@ class ModelManager:
             futures = [executor.submit(self.load_model, config) for config in model_configs]
             models = []
             for future in as_completed(futures):
-                try:
-                    model = future.result()
+                model = future.result()
+                if model:
                     models.append(model)
-                except Exception:
-                    pass
         
         self.models = models
         self.loaded = True
@@ -84,7 +85,6 @@ class ChatRequest(BaseModel):
     top_p: float = 0.95
     temperature: float = 0.7
 
-@spaces.GPU(duration=0)
 def generate_chat_response(request, model_data):
     try:
         user_input = normalize_input(request.message)
@@ -138,8 +138,8 @@ def generate_chat(request: ChatRequest):
             try:
                 response = future.result()
                 responses.append(response)
-            except Exception:
-                pass
+            except Exception as e:
+                print(f"Error in response generation: {e}")
     
     if not responses:
         raise HTTPException(status_code=500, detail="Error: No responses generated.")