# FR - tell us what model 'auto' is using

**URL:** <https://forum.cursor.com/t/fr-tell-us-what-model-auto-is-using/87491>\
**Category:** Feature Requests\
**Created:** [May 7, 2025, 10:06am UTC](https://forum.cursor.com/t/fr-tell-us-what-model-auto-is-using/87491 "2025-05-07T10:06:57Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![mwjt42](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/mwjt42/32/7270_2.png) [@mwjt42](https://forum.cursor.com/u/mwjt42)\
**Post date:** [May 7, 2025, 10:06am UTC](https://forum.cursor.com/t/fr-tell-us-what-model-auto-is-using/87491/1 "2025-05-07T10:06:57Z")

</div>

Just a nice bit of UI feedback, but I’d be interested to see which model auto chooses.

---

<div class="post-metadata">

**Author:** ![CasablancA](https://avatars.discourse-cdn.com/v4/letter/c/3da27b/32.png) [@CasablancA](https://forum.cursor.com/u/CasablancA)\
**Post date:** [May 7, 2025, 4:43pm UTC](https://forum.cursor.com/t/fr-tell-us-what-model-auto-is-using/87491/2 "2025-05-07T16:43:23Z")

</div>

from its reply, I think it’s GPT-4.1

---

<div class="post-metadata">

**Author:** ![liquefy](https://avatars.discourse-cdn.com/v4/letter/l/edb3f5/32.png) [@liquefy](https://forum.cursor.com/u/liquefy)\
**Post date:** [May 7, 2025, 4:56pm UTC](https://forum.cursor.com/t/fr-tell-us-what-model-auto-is-using/87491/3 "2025-05-07T16:56:30Z")

</div>

+1 valid FR

---

<div class="post-metadata">

**Author:** ![kejo84](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/kejo84/32/69971_2.png) [@kejo84](https://forum.cursor.com/u/kejo84)\
**Post date:** [July 10, 2025, 8:26pm UTC](https://forum.cursor.com/t/fr-tell-us-what-model-auto-is-using/87491/4 "2025-07-10T20:26:24Z")

</div>

They also use gpt-4.1-mini with the auto mode.

---

<div class="post-metadata">

**Author:** ![Priders\_zyh](https://avatars.discourse-cdn.com/v4/letter/p/f19dbf/32.png) [@Priders\_zyh](https://forum.cursor.com/u/Priders_zyh)\
**Post date:** [July 15, 2025, 5:09am UTC](https://forum.cursor.com/t/fr-tell-us-what-model-auto-is-using/87491/5 "2025-07-15T05:09:10Z")

</div>

You are currently having a conversation with me (GPT-4.0/4.1 series), an AI assistant developed by OpenAI and currently running in the Cursor development environment. My underlying model is GPT-4.1, which is specifically optimized for code understanding, generation and collaboration. If you have any questions about code, project structure or development, feel free to ask me at any time!

---

<div class="post-metadata">

**Author:** ![mwjt42](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/mwjt42/32/7270_2.png) [@mwjt42](https://forum.cursor.com/u/mwjt42)\
**Post date:** [July 15, 2025, 6:34am UTC](https://forum.cursor.com/t/fr-tell-us-what-model-auto-is-using/87491/6 "2025-07-15T06:34:16Z")

</div>

That’s missing the point. The auto model varies, we’re requesting a small bit of UI feedback to inform the user of which model was chosen, per request.

---

<div class="post-metadata">

**Author:** ![RafeSacks](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/rafesacks/32/15056_2.png) [@RafeSacks](https://forum.cursor.com/u/RafeSacks)\
**Post date:** [July 25, 2025, 12:42am UTC](https://forum.cursor.com/t/fr-tell-us-what-model-auto-is-using/87491/7 "2025-07-25T00:42:43Z")

</div>

Auto seems pretty bad, often costing me time. I’m trying other models directly simply because I don’t trust auto any more and I don’t know what model to blame.