# Cursor Has Become Fast for Slow Requests

**URL:** <https://forum.cursor.com/t/cursor-has-become-fast-for-slow-requests/45695>\
**Category:** Feedback\
**Created:** [January 28, 2025, 6:40pm UTC](https://forum.cursor.com/t/cursor-has-become-fast-for-slow-requests/45695 "2025-01-28T18:40:06Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![andresreibel](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/andresreibel/32/4864_2.png) [@andresreibel](https://forum.cursor.com/u/andresreibel)\
**Post date:** [January 28, 2025, 6:40pm UTC](https://forum.cursor.com/t/cursor-has-become-fast-for-slow-requests/45695/1 "2025-01-28T18:40:06Z")

</div>

I have been in slow requests for a couple of days and they are fast. If it stays fast AND you add DeepSeek R1 to agent composer (via API Key) I will consider swapping back from VSCode + Cline + DeepSeek R1

---

<div class="post-metadata">

**Author:** ![danperks](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/danperks/32/19087_2.png) [@danperks](https://forum.cursor.com/u/danperks)\
**Post date:** [January 30, 2025, 1:22pm UTC](https://forum.cursor.com/t/cursor-has-become-fast-for-slow-requests/45695/2 "2025-01-30T13:22:41Z")

</div>

Hey,

The slow queue is really at it’s slowest with the Anthropic models, as we are limited on the capacity we receive from Antrhopic during peak times.

However, as you have found, the other models including DeepSeek and OpenAI models should work with much shorter queues!

We are adding DeepSeek as an agent model soon, the team is actively working on it at the moment, so expect to see that added soon!

---

<div class="post-metadata">

**Author:** ![utku](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/utku/32/20061_2.png) [@utku](https://forum.cursor.com/u/utku)\
**Post date:** [January 30, 2025, 1:34pm UTC](https://forum.cursor.com/t/cursor-has-become-fast-for-slow-requests/45695/3 "2025-01-30T13:34:00Z")

</div>

this is great to hear!

---

<div class="post-metadata">

**Author:** ![andresreibel](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/andresreibel/32/4864_2.png) [@andresreibel](https://forum.cursor.com/u/andresreibel)\
**Post date:** [January 30, 2025, 3:27pm UTC](https://forum.cursor.com/t/cursor-has-become-fast-for-slow-requests/45695/4 "2025-01-30T15:27:10Z")

</div>

Thanks, that’s amazing, especially if you can bring DeepSeek R1 to Composer (agent) asap.

Question: which DeepSeek R1 model (number of parameters) are you working with and can we use an API key to hook up our own version of DeepSeek R1?

Appreciate it!

---

<div class="post-metadata">

**Author:** ![Raphael909](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/raphael909/32/20893_2.png) [@Raphael909](https://forum.cursor.com/u/Raphael909)\
**Post date:** [January 31, 2025, 1:20am UTC](https://forum.cursor.com/t/cursor-has-become-fast-for-slow-requests/45695/5 "2025-01-31T01:20:51Z")

</div>

there you go -

> [@Which version of Deepseek R1 does Cursor use?](https://forum.cursor.com/t/which-version-of-deepseek-r1-does-cursor-use/45703/2):
>
> Hey, we are using the DeepSeek R1 model with 671B parameters.

---

<div class="post-metadata">

**Author:** ![system](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/system/32/99069_2.png) [@system](https://forum.cursor.com/u/system)\
**Post date:** [June 18, 2026, 10:26am UTC](https://forum.cursor.com/t/cursor-has-become-fast-for-slow-requests/45695/6 "2026-06-18T10:26:11Z")

</div>

Closed on June 18, 2026, 10:26am UTC