Update app.py

app.py

@@ -1,6 +1,7 @@
 import streamlit as st
 import torch
 import threading
+from peft import PeftModel
 from transformers import (
     AutoModelForCausalLM,
     AutoTokenizer,
@@ -25,18 +26,23 @@ def load_model():
         trust_remote_code=True
     )
 
-    model = AutoModelForCausalLM.from_pretrained(
+    base_model = AutoModelForCausalLM.from_pretrained(
         MODEL_ID,
         torch_dtype=torch.float32,
         device_map=None
     )
 
+    try:
+        model = PeftModel.from_pretrained(base_model, MODEL_ID)
+        print("✅ LoRA adapter loaded")
+    except Exception:
+        model = base_model
+        print("ℹ️ No LoRA adapter found, using base model")
+
     model.to("cpu")
     model.eval()
     return tokenizer, model
 
-tokenizer, model = load_model()
-
 # -------- SESSION STATE --------
 if "history" not in st.session_state:
     st.session_state.history = []