Introducing Cursor 2.0. Our first coding model and the best way to code with agents.

Oct 29, 2025 · 4:12 PM UTC

568
1,022
8,099
3,571,687
Composer is a frontier coding model that completes tasks in under 30 seconds.
22
38
643
489,961
Easily run the same prompt across models for maximum intelligence.
8
10
327
121,748
With a built-in browser, agents can now run and test their code.
9
13
413
79,233
Turn speech into code with voice mode.
20
11
371
114,274
Read about everything that's new in 2.0. More tomorrow! cursor.com/blog/2-0

Introducing Cursor 2.0 and Composer · Cursor

A new interface and our first coding model, both purpose-built for working with agents.

cursor.com
17
34
495
284,965
Sort replies: Relevant Recent Liked
Replying to @cursor_ai
brb,
6
4
254
18,602
Replying to @cursor_ai
holy shit!
1
47
4,222
Replying to @cursor_ai
We would love to benchmark this using coding benchmarks that we run like-for-like across all models. Let us know!
2
46
6,696
Replying to @cursor_ai
Huge upgrade! 👀👀👀
26
3,518
Replying to @cursor_ai
First test drive. Its fast, then it deleted the entire file without telling me . great
1
28
5,129
Replying to @cursor_ai
Congrats on the launch!
1
19
2,693
Replying to @cursor_ai
excited to use and criticize heavily!
1
18
2,208
Replying to @cursor_ai
So fast, won't let me doom scroll on Twitter at all. Congratulations on the release!
12
7,532
Replying to @cursor_ai
Y’all cooking. I had the opportunity to test Composer 1 on a difficult task and it performed way better than all the LLMs I tried.
11
4,103
Replying to @cursor_ai
Read our full vibe check on the new release @every every.to/vibe-check/vibe-che…

Vibe Check: Cursor 2.0 and Composer 1 Alpha

Two new things: A code editor designed to manage agents and a lightning-fast model

every.to
1
2
9
1,938
Replying to @cursor_ai
I‘ve been using @cursor_ai 2.0 in the early access program and I love it! Favorite features: - Agent code review - Voice input - Integrated browser - with ability to select HTML elements as context - Multi model tasks - give same task to different models and compare
1
9
4,753
Replying to @cursor_ai
Wait, so does Composer use tokens from all three models? Isn’t that quite wasteful?
4
9
25,074
Replying to @cursor_ai
yo what
9
940
Replying to @cursor_ai
🤯
6
1,155
Replying to @cursor_ai
only app people love updating!
6
1,407
Replying to @cursor_ai
Let’s go!
5
1,211
Replying to @cursor_ai
Congratulations!! You are on fire. 🔥
3
5,127
Replying to @cursor_ai
Bang
4
1,664
Replying to @cursor_ai
new toy just dropped
1
4
914
Replying to @cursor_ai
W cursor
1
4
916
Replying to @cursor_ai
Amazing launch, congrats to everyone!
3
828
Replying to @cursor_ai
Congrats for the release! Happy to test it!
2
921
Replying to @cursor_ai
Every dev when they saw the browser part
2
1,578
Replying to @cursor_ai
oh snap. y’all cooked!
2
857
Replying to @cursor_ai
Colf spotted 👀
1
585
Replying to @cursor_ai
Amazing!
2
76
Replying to @cursor_ai
wow this is hugee. looks fast
2
457
Replying to @cursor_ai
ok wowowowow
2
540
Replying to @cursor_ai
Woah!
1
272
Replying to @cursor_ai
Congrats on the release. It’s unfortunate that I have stay on v1.6
Alireza
@alireza_hadjar
29 Oct 2025
Replying to @milichab @cursor_ai
Just tried out v2.0.34 and the same issue happens. As you can see in the video, everything is unresponsive and laggy and the cpu utilization is at 225% @cursor_ai
1
480
Replying to @cursor_ai
👀 anyone tried it? Worth using over Claude code or ChatGPT codex extension?
885
Load more