# Continuous execution lead to decreasing inference time

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcontinuous-execution-lead-to-decreasing-inference-time%2F114317%3E%5C
**Category:** Beginners\
**Created:** [October 28, 2024, 8:55am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcontinuous-execution-lead-to-decreasing-inference-time%2F114317 "2024-10-28T08:55:44Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![allenwang37](https://avatars.discourse-cdn.com/v4/letter/a/bbe5ce/32.png) [@allenwang37](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fallenwang37)%5C
**Post date:** [October 28, 2024, 8:55am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcontinuous-execution-lead-to-decreasing-inference-time%2F114317%2F1 "2024-10-28T08:55:44Z")

</div>

> for i in range(50):  
> model.generate()  
> I found the inference time is decreasing. I guess it is relative to cuda cache or something. I want to know what factors influenced this result？