Update app.py

app.py

@@ -2,9 +2,20 @@ import gradio as gr
 
 # Load your model and tokenizer
 from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
 
-model_name = "ahmedbasemdev/LLama3.2-fine-tuned"  # Replace with your model name
-model = AutoModelForCausalLM.from_pretrained(model_name)
+# Specify the model name
+model_name = "ahmedbasemdev/llama-3.2-3b-ChatBot"
+
+# Load the model with 8-bit quantization
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    device_map="auto",  # Automatically map the model to the available device (CPU)
+    load_in_8bit=True,  # Enable 8-bit quantization
+    torch_dtype=torch.float16  # Use mixed precision
+)
+
+# Load the tokenizer
 tokenizer = AutoTokenizer.from_pretrained(model_name)
 
 def single_inference(question):