# Open API standard for open-source LLMs

**URL:** <https://discuss.huggingface.co/t/open-api-standard-for-open-source-llms/45241>\
**Category:** Research\
**Created:** [July 1, 2023, 9:47am UTC](https://discuss.huggingface.co/t/open-api-standard-for-open-source-llms/45241 "2023-07-01T09:47:41Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![imone](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/imone/32/17270_2.png) [@imone](https://discuss.huggingface.co/u/imone)\
**Post date:** [July 1, 2023, 9:47am UTC](https://discuss.huggingface.co/t/open-api-standard-for-open-source-llms/45241/1 "2023-07-01T09:47:41Z")

</div>

Does anyone have experience/interest in creating API standards? I think we need an API standard for open source models, like OpenAI Completions/ChatCompletions. This will greatly simplify running benchmarks and evaluations on open-source models, since we don’t need to implement inference code/dialogue templates for each model.

Personally, I use a custom-written local OpenAI-compatible API server to run standard benchmarks. This allows me to easily run benchmark code from different sources just by modifying api\_base, since almost all benchmarks support OpenAI models.