Update app.py

app.py

@@ -2,7 +2,8 @@
 import torch
 import pandas as pd
 import streamlit as st
-from langchain.llms import HuggingFaceLLM
+from langchain.llms import HuggingFacePipeline
+from transformers import pipeline
 from huggingface_hub import login
 
 # Tu token secreto de Hugging Face
@@ -11,7 +12,9 @@ login(huggingface_token)
 
 # Cargar el modelo Llama 3.1
 model_name = "meta-llama/llama-3.1-8b-instruct"  # Asegúrate de que este sea el nombre correcto del modelo
-llm = HuggingFaceLLM(model_name=model_name, torch_dtype=torch.float16)
+# Usar transformers pipeline para carga
+llm_pipeline = pipeline("text-generation", model=model_name, device=0 if torch.cuda.is_available() else -1)
+llm = HuggingFacePipeline(pipeline=llm_pipeline)
 
 # Interfaz de Streamlit
 st.title("Cosine Similarity con Llama 3.1")