upd

src/streamlit_app.py

@@ -1,5 +1,5 @@
 import streamlit as st
-from transformers import AutoTokenizer, AutoModelForSeq2SeqLM, pipeline
+from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
 import os
 
 # Disable Streamlit usage stats to avoid permission issues
@@ -16,8 +16,8 @@ def load_model():
     os.makedirs(cache_dir, exist_ok=True)
     
     tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct", cache_dir=cache_dir)
-    model = AutoModelForSeq2SeqLM.from_pretrained("tianzhechu/BookQA-7B-Instruct", cache_dir=cache_dir)
-    return pipeline("text2text-generation", model=model, tokenizer=tokenizer)
+    model = AutoModelForCausalLM.from_pretrained("tianzhechu/BookQA-7B-Instruct", cache_dir=cache_dir)
+    return pipeline("text-generation", model=model, tokenizer=tokenizer)
 
 st.set_page_config(page_title="LLM Demo", layout="centered")
 st.title("🚀 FLAN-T5 Small - HuggingFace Demo")