Update app.py

app.py

@@ -56,8 +56,8 @@ def get_llama_response(query):
     # response = tokenizer.decode(outputs[0]["generated_text"], skip_special_tokens=True)
     pipe = pipeline(
         task="text-generation",
-        model=model,
-        tokenizer=tokenizer,
+        model=model=st.session_state.llama_model,
+        tokenizer=st.session_state.llama_tokenizer,
         max_new_tokens=128,
         return_full_text=False,
     )