# The vllm/vllm-openai version 0.9.1 is nearly 30% faster compared to lmsysorg/sglang:v0.4.7.post, but it stops running every two to three hours

**URL:** <https://discuss.vllm.ai/t/the-vllm-vllm-openai-version-0-9-1-is-nearly-30-faster-compared-to-lmsysorg-sglang-v0-4-7-post-but-it-stops-running-every-two-to-three-hours/898>\
**Category:** DeepSeek\
**Created:** [June 23, 2025, 12:51pm UTC](https://discuss.vllm.ai/t/the-vllm-vllm-openai-version-0-9-1-is-nearly-30-faster-compared-to-lmsysorg-sglang-v0-4-7-post-but-it-stops-running-every-two-to-three-hours/898 "2025-06-23T12:51:02Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![start-dzl](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.vllm.ai/start-dzl/32/404_2.png) [@start-dzl](https://discuss.vllm.ai/u/start-dzl)\
**Post date:** [June 23, 2025, 12:51pm UTC](https://discuss.vllm.ai/t/the-vllm-vllm-openai-version-0-9-1-is-nearly-30-faster-compared-to-lmsysorg-sglang-v0-4-7-post-but-it-stops-running-every-two-to-three-hours/898/1 "2025-06-23T12:51:02Z")

</div>

[**bug**](https://github.com/vllm-project/vllm/issues/19933)