Update llm_model.py

llm_model.py

@@ -66,4 +66,10 @@ async def generate_response(text, app_config):
         if start != -1:
             decoded = decoded[start + len(tag):].strip()
             break
-    return decoded, top_conf
+    return decoded, top_conf
+
+def get_model():
+    return model
+
+def get_tokenizer():
+    return tokenizer