# Question - toolUsePromptTokenCount

**URL:** <https://discuss.ai.google.dev/t/question-tooluseprompttokencount/137102>\
**Category:** Gemini API\
**Tags:** api, gemini\
**Created:** [March 30, 2026, 9:00pm UTC](https://discuss.ai.google.dev/t/question-tooluseprompttokencount/137102 "2026-03-30T21:00:46Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Bmcneko](https://d1dlmcr85iqnpo.cloudfront.net/user_avatar/discuss.ai.google.dev/bmcneko/32/32994_2.png) [@Bmcneko](https://discuss.ai.google.dev/u/Bmcneko)\
**Post date:** [March 30, 2026, 9:00pm UTC](https://discuss.ai.google.dev/t/question-tooluseprompttokencount/137102/1 "2026-03-30T21:00:46Z")

</div>

Good morning! I’ve recently started integrating the Gemini API into a few projects, and I have a question. When using tools like Google Search, I notice the following metadata: `toolUsePromptTokenCount`. Is there a way to limit how many tokens this tool consumes? It’s consistently quite high. Additionally, is it possible to restrict the number of searches the model attempts to perform?