Pytorch and GPU

app.py

@@ -1,12 +1,15 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
 from transformers import pipeline
+import torch
 
 """
 For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
 """
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-pipe = pipeline("text-generation", model="HiTZ/latxa-7b-v1")
+pipe = pipeline("text-generation", model="HiTZ/latxa-7b-v1", device=0)
+
 
 def respond(
     message,