# Is there a way to connect my own models to use in cursor

**URL:** <https://forum.cursor.com/t/is-there-a-way-to-connect-my-own-models-to-use-in-cursor/131859>\
**Category:** Help\
**Created:** [August 25, 2025, 10:52am UTC](https://forum.cursor.com/t/is-there-a-way-to-connect-my-own-models-to-use-in-cursor/131859 "2025-08-25T10:52:25Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![oly99](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/oly99/32/77392_2.png) [@oly99](https://forum.cursor.com/u/oly99)\
**Post date:** [August 25, 2025, 10:52am UTC](https://forum.cursor.com/t/is-there-a-way-to-connect-my-own-models-to-use-in-cursor/131859/1 "2025-08-25T10:52:25Z")

</div>

is there a way to connect my own models to use in cursor i have lm studio with some models in esspecialy claude 3.7 when the agent has got really stuck ive asked my model and it gave me a working solution streight away but its a pain cus i have to save the problem code to file and do a promt etc is there a way i can add it cursor would be better. or would using vs studio code with lm studio linked be a better way . and run the code though that and then generate a prompt to run on cursor i prob would not want to edit the code in vs studio code but just to find the bug and make the fix prompt.

---

<div class="post-metadata">

**Author:** ![deanrie](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/deanrie/32/1263_2.png) [@deanrie](https://forum.cursor.com/u/deanrie)\
**Post date:** [August 25, 2025, 11:29am UTC](https://forum.cursor.com/t/is-there-a-way-to-connect-my-own-models-to-use-in-cursor/131859/2 "2025-08-25T11:29:42Z")

</div>

Hey, local models are not supported, but you can check out this topic for more information:

> [@Support local LLM's](https://forum.cursor.com/t/support-local-llms/1099):
>
> Hi There, I’m wondering if there are plans in the future to support local LLM’s within Cursor? While today you support GPT-3.5 & GPT-4, it would be great if we could point Cursor to a local LLM on the machine that has been specifically tuned on a particular codebase(s).

---

<div class="post-metadata">

**Author:** ![oly99](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/oly99/32/77392_2.png) [@oly99](https://forum.cursor.com/u/oly99)\
**Post date:** [August 25, 2025, 3:40pm UTC](https://forum.cursor.com/t/is-there-a-way-to-connect-my-own-models-to-use-in-cursor/131859/4 "2025-08-25T15:40:30Z")

</div>

i think there is a app that adds a api end point that cursor can connect to havent tried it though

---

<div class="post-metadata">

**Author:** ![deanrie](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/deanrie/32/1263_2.png) [@deanrie](https://forum.cursor.com/u/deanrie)\
**Post date:** [August 25, 2025, 4:00pm UTC](https://forum.cursor.com/t/is-there-a-way-to-connect-my-own-models-to-use-in-cursor/131859/5 "2025-08-25T16:00:12Z")

</div>

You can try [OpenRouter](https://openrouter.ai/), it has many models, including free ones.