# Inference API offline model limit

**URL:** <https://discuss.huggingface.co/t/inference-api-offline-model-limit/18118>\
**Category:** 🤗Transformers\
**Created:** [May 18, 2022, 7:20pm UTC](https://discuss.huggingface.co/t/inference-api-offline-model-limit/18118 "2022-05-18T19:20:08Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![aliosm](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/aliosm/32/4814_2.png) [@aliosm](https://discuss.huggingface.co/u/aliosm)\
**Post date:** [May 18, 2022, 7:20pm UTC](https://discuss.huggingface.co/t/inference-api-offline-model-limit/18118/1 "2022-05-18T19:20:08Z")

</div>

Hello team,

I’m uaing the inference API to create a simple website, I’m using the pro plan and I want to know how many minutes the API keeps the model in memory before off-loading it?

For example, when I request the model the first time, the inference API puts the model on memory, then after how many minutes it will be off-loaded?

Thanks!

---

<div class="post-metadata">

**Author:** ![chin-cyber](https://avatars.discourse-cdn.com/v4/letter/c/ad7895/32.png) [@chin-cyber](https://discuss.huggingface.co/u/chin-cyber)\
**Post date:** [May 2, 2024, 10:35am UTC](https://discuss.huggingface.co/t/inference-api-offline-model-limit/18118/2 "2024-05-02T10:35:48Z")

</div>

it is stored permanently into your cache, i will suggest take it out and host it on your cloud instance