Update src/rag.py

src/rag.py

@@ -479,7 +479,7 @@ def load_generation_model():
     model_name = "meta-llama/Llama-3.2-3B-Instruct"  # ~1.1B parameters
     
     import os
-    token = os.getenv('HF_TOKEN'))
+    token = os.getenv('HF_TOKEN')
     
     try:
         # First try loading with 4-bit quantization if bitsandbytes is available