# Inference speed

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Finference-speed%2F55325%3E%5C
**Category:** Spaces\
**Created:** [September 17, 2023, 10:41am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Finference-speed%2F55325 "2023-09-17T10:41:48Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![m-ali01](https://avatars.discourse-cdn.com/v4/letter/m/4da419/32.png) [@m-ali01](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fm-ali01)%5C
**Post date:** [September 17, 2023, 10:41am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Finference-speed%2F55325%2F1 "2023-09-17T10:41:48Z")

</div>

Hello,

Does anyone have a benchmark of inference speed with llama-2 using huggingface spaces?

Another related question: are models hosted through the huggingface inference endpoint result in higher tokens/s than hf spaces using the same gpu?

Thanks