# What is the context window for GPT-4 when using GPTs?

**URL:** <https://community.openai.com/t/what-is-the-context-window-for-gpt-4-when-using-gpts/581352>\
**Category:** Plugin store\
**Tags:** plugin-development, gpt-development\
**Created:** [January 10, 2024, 7:45pm UTC](https://community.openai.com/t/what-is-the-context-window-for-gpt-4-when-using-gpts/581352 "2024-01-10T19:45:34Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![danfhern](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/danfhern/32/243394_2.png) [@danfhern](https://community.openai.com/u/danfhern)\
**Post date:** [January 10, 2024, 7:45pm UTC](https://community.openai.com/t/what-is-the-context-window-for-gpt-4-when-using-gpts/581352/1 "2024-01-10T19:45:34Z")

</div>

Simple question: what is the context window when creating custom GPTs?

From this thread it looks like it may be \< 32k for all GPT-4 not in teams plan? [https://twitter.com/aidan\_mclau/status/1745145277961396716](https://twitter.com/aidan_mclau/status/1745145277961396716)

It would be good to know exactly what it is so it can be accounted for designing how many tokens to return back from actions.

I.e. we are building [https://gpt.videoinsights.ai](https://gpt.videoinsights.ai) and it’s important that we know what the upper limit is because it can very easily be hit with long video transcripts.