Update app.py

app.py

@@ -26,21 +26,23 @@ app = FastAPI(
 async def health_check():
     return {"status": "healthy"}
 
-# Load T5 model from Hub
+# Load T5 model from local cache
 T5_REPO = "MGZON/mgzon-flan-t5-base"
 try:
-    logger.info(f"Loading tokenizer for {T5_REPO} with HF_TOKEN")
+    logger.info(f"Loading tokenizer for {T5_REPO} from local cache")
     t5_tokenizer = AutoTokenizer.from_pretrained(
         T5_REPO,
         cache_dir=CACHE_DIR,
-        use_auth_token=os.environ.get("HF_TOKEN")
+        use_auth_token=os.environ.get("HF_TOKEN"),
+        local_files_only=True
     )
     logger.info(f"Successfully loaded tokenizer for {T5_REPO}")
-    logger.info(f"Loading model for {T5_REPO}")
+    logger.info(f"Loading model for {T5_REPO} from local cache")
     t5_model = AutoModelForSeq2SeqLM.from_pretrained(
         T5_REPO,
         cache_dir=CACHE_DIR,
-        use_auth_token=os.environ.get("HF_TOKEN")
+        use_auth_token=os.environ.get("HF_TOKEN"),
+        local_files_only=True
     )
     logger.info(f"Successfully loaded model for {T5_REPO}")
 except Exception as e:
@@ -101,4 +103,4 @@ def ask(req: AskRequest):
         return {"model": model_name, "response": answer}
     except Exception as e:
         logger.error(f"Error processing request: {str(e)}")
-        raise HTTPException(status_code=500, detail=f"خطأ أثناء معالجة الطلب: {str(e)}")
+        raise HTTPException(status_code=500, detail=f"خطأ أثناء معالجة الطلب: {str(e)}")