Update app.py

app.py

@@ -50,8 +50,10 @@ def load_llm():
     if model_instance is None:
         print("Cargando el modelo...")
         model_instance = AutoModelForCausalLM.from_pretrained(
-            "TheBloke/Llama-2-7B-Chat-GGUF",
-            model_file='llama-2-7b-chat.Q5_K_M.gguf',
+            #"TheBloke/Llama-2-7B-Chat-GGUF",
+            "TheBloke/phi-2-GGUF",
+            #model_file='llama-2-7b-chat.Q5_K_M.gguf',
+            model_file='phi-2.Q4_K_M.gguf',
             model_type='llama',
             max_new_tokens=2048,
             context_length = 4096,