The Battle of LLMs: A Comparative Study in Conversational QA Tasks