# Serve the same model replicas on the same GPU

**URL:** <https://discuss.ray.io/t/serve-the-same-model-replicas-on-the-same-gpu/14772>\
**Category:** Ray Serve\
**Created:** [May 23, 2024, 5:30am UTC](https://discuss.ray.io/t/serve-the-same-model-replicas-on-the-same-gpu/14772 "2024-05-23T05:30:05Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![islam\_almersawi](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/islam_almersawi/32/6155_2.png) [@islam\_almersawi](https://discuss.ray.io/u/islam_almersawi)\
**Post date:** [May 23, 2024, 5:30am UTC](https://discuss.ray.io/t/serve-the-same-model-replicas-on-the-same-gpu/14772/1 "2024-05-23T05:30:05Z")

</div>

Is it possible to deploy multiple replicas of the same model on the same GPU? or it’s a must to allocate a different GPU for each replica?