Update app.py

app.py

@@ -1,10 +1,21 @@
 import gradio as gr
-from transformers import pipeline
+from transformers import pipeline, AutoModelForCausalLM, AutoTokenizer
 
+# Load model with 8-bit precision
+model_name = "yasserrmd/SmolLM2-156M-synthetic-dlp"
+model = AutoModelForCausalLM.from_pretrained(
+    model_name, 
+    load_in_8bit=True,  # Enable 8-bit precision
+    device_map="auto"   # Automatically allocate to GPU/CPU
+)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+# Load the pipeline
 generator = pipeline(
-        "text-generation",
-        model="yasserrmd/SmolLM2-156M-synthetic-dlp"
-    )
+    "text-generation",
+    model=model,
+    tokenizer=tokenizer
+)
 
 def chat_assistant(chat_history, user_input):
     """Generate a response based on user input and chat history."""