There are reports that local large language models (LLMs) may not be useful if they cannot interact with external tools, regardless of their size or parameters. Some users have shared their experiences with running LLMs locally, with mixed results, including replacing paid apps and encountering limitations. The effectiveness and reliability of local LLMs compared to cloud-based alternatives, such as ChatGPT, remains a topic of discussion.