Update app.py

app.py

@@ -1,5 +1,6 @@
 import os
 import gradio as gr
+import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM
 
 access_token = os.getenv('HF_TOKEN')
@@ -10,7 +11,13 @@ access_token = "your_access_token_here"
 
 # Load the tokenizer and model from the Hugging Face repository
 tokenizer = AutoTokenizer.from_pretrained(repo_id, token=access_token)
-model = AutoModelForCausalLM.from_pretrained(repo_id, token=access_token)
+
+model = AutoModelForCausalLM.from_pretrained(
+    repo_id, 
+    token=access_token, 
+    torch_dtype=torch.bfloat16,  # or use torch.bfloat16 if supported
+    device_map="auto"  # Automatically use available GPU/CPU efficiently
+)
 
 def respond(
     message,