Update app.py

app.py

@@ -5,13 +5,7 @@ from langchain_community.embeddings import HuggingFaceEmbeddings
 from langchain_community.document_loaders import PyMuPDFLoader
 from langchain_text_splitters import CharacterTextSplitter
 from langchain.chains import RetrievalQA
-from langchain_huggingface import HuggingFaceEndpoint  # Updated import
-from huggingface_hub import login
-
-# Authentication
-if not os.environ.get('HF_TOKEN'):
-    raise ValueError("❌ Add HF_TOKEN in Space secrets!")
-login(token=os.environ.get('HF_TOKEN'))
+from transformers import pipeline  # Local model execution
 
 def create_qa_system():
     try:
@@ -40,19 +34,19 @@ def create_qa_system():
         # Build vector store
         db = FAISS.from_documents(texts, embeddings)
         
-        # Initialize LLM with corrected parameters
-        llm = HuggingFaceEndpoint(
-            endpoint_url="https://api-inference.huggingface.co/models/google/flan-t5-small",
-            task="text2text-generation",
-            temperature=0.2,  # Direct parameter
-            max_new_tokens=128,  # Correct parameter name
-            huggingfacehub_api_token=os.environ.get('HF_TOKEN')
+        # Local model pipeline
+        qa_pipeline = pipeline(
+            "text2text-generation",
+            model="google/flan-t5-small",  # Runs locally
+            device=-1,  # Use CPU
+            max_length=128,
+            temperature=0.2
         )
         
         return RetrievalQA.from_chain_type(
-            llm=llm,
+            llm=qa_pipeline,
             chain_type="stuff",
-            retriever=db.as_retriever(search_kwargs={"k": 2}))
+            retriever=db.as_retriever(search_kwargs={"k": 2})
     except Exception as e:
         raise gr.Error(f"Initialization failed: {str(e)}")