# Chosing different models for different groups/tiers

**URL:** <https://meta.discourse.org/t/chosing-different-models-for-different-groups-tiers/410660>\
**Category:** Feature\
**Tags:** ai\
**Created:** [August 21, 2026, 9:26pm UTC](https://meta.discourse.org/t/chosing-different-models-for-different-groups-tiers/410660 "2026-08-21T21:26:45Z")\
**Posts on this page:** 1\
**Showing post:** 1

<div class="post-metadata">

**Author:** ![satonotdead](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/satonotdead/32/447830_2.png) [@satonotdead](https://meta.discourse.org/u/satonotdead)\
**Post date:** [August 21, 2026, 9:26pm UTC](https://meta.discourse.org/t/chosing-different-models-for-different-groups-tiers/410660/1 "2026-08-21T21:26:45Z")

</div>

I know we can limit the AI usage of our Agents through different types of quotas.

I wonder if it could be good to have a distinction about **what models each group can use** in the same Agents.

We want to preserve expensive tokens for TL2/3 but also allow a less expensive model for TL1 users.

---

_[View the full topic](https://meta.discourse.org/t/chosing-different-models-for-different-groups-tiers/410660)._