# Minor UI bug in LLM Usage page

**URL:** https://meta.discourse.org/t/minor-ui-bug-in-llm-usage-page/391273
**Category:** UX
**Tags:** pr-welcome, ai
**Created:** [December 14, 2025, 12:36pm UTC](https://meta.discourse.org/t/minor-ui-bug-in-llm-usage-page/391273 "2025-12-14T12:36:27Z")
**Posts on this page:** 1
**Showing post:** 1

<div class="post-metadata">

### Author: ![RBoy](https://avatars.discourse-cdn.com/v4/letter/r/2bfe46/32.png) [@RBoy](https://meta.discourse.org/u/RBoy)
#### Post date: [December 14, 2025, 12:36pm UTC](https://meta.discourse.org/t/minor-ui-bug-in-llm-usage-page/391273/1 "2025-12-14T12:36:27Z")

</div>

A minor bug on the LLM usage page. When you create a new LLM model and run a test for it, it shows up in the usage report under llm\_validator, but since there’s no name of the llm as yet the model name shows up in the filter list (no LLM name), in this case `moonshotai`

 ![image](https://global.discourse-cdn.com/meta/original/4X/9/7/9/979faae5c7499775ef7059c35d5a394993ced7d3.png)

However after selecting it, the dropdown filter changes to 0 instead of `moonshotai` and all the statistics show up as 0

 ![image](https://global.discourse-cdn.com/meta/original/4X/e/0/8/e0838b0f57b7bccf65fc4c83a5330178b49aef50.png)

---

_[View the full topic](https://meta.discourse.org/t/minor-ui-bug-in-llm-usage-page/391273)._