Update app.py

app.py

@@ -42,6 +42,9 @@ model = transformers.AutoModelForCausalLM.from_pretrained(
     #quantization_config=bnb_config,
     device_map='auto')
 
+# Offload specific parts of the model to disk
+model = transformers.disk_offload(model, "Enron_case_RAG")
+
 tokenizer = AutoTokenizer.from_pretrained(model_id)
 
 ##############################################################################