# MemGPT and function calling

**URL:** <https://community.deeplearning.ai/t/memgpt-and-function-calling/478627>\
**Category:** Functions, Tools and Agents with LangChain\
**Created:** [October 27, 2023, 9:58am UTC](https://community.deeplearning.ai/t/memgpt-and-function-calling/478627 "2023-10-27T09:58:58Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Robolicious](https://avatars.discourse-cdn.com/v4/letter/r/ec9cab/32.png) [@Robolicious](https://community.deeplearning.ai/u/Robolicious)\
**Post date:** [October 27, 2023, 9:58am UTC](https://community.deeplearning.ai/t/memgpt-and-function-calling/478627/1 "2023-10-27T09:58:58Z")

</div>

While I was doing this short course I was reading the MemGPT paper.  
Do I understand correctly that when we use langchaing and OpenAI’s function calling + langchain’s memory feature we already use the “external context”?

In that sense, what is new then with MemGPT? What am I missing?