changing to llama 3.2 1B

app.py

@@ -3,14 +3,17 @@ from transformers import pipeline
 import os
 
 access_token = os.getenv("Token")
-pipe = pipeline("text-generation", model="google/gemma-2-2b-it", token=access_token)
+pipe = pipeline("text-generation", model="meta-llama/Llama-3.2-1B-Instruct", token=access_token)
 
-def predict(query):
-    return pipe(query)
+def predict(system="you are a helpful assistant", query):
+    messages = []
+    messages.append({"role": "system","content": system})
+    messages.append({"role": "system","content": query})
+    return pipe(messages,max_new_tokens=256)[0]["generated_text"][-1]
 
 gradio_app = gr.Interface(
     predict,
-    inputs="text", outputs="text",
+    inputs=["text","text"], outputs="text",
     title="Gemma 2B",
 )