Cursor Composer
Cursor's in-house coding model for agentic tasks inside the Cursor AI code editor.
HN commenters use it as a fast, cheap daily workhorse123 for small and routine coding tasks,456 often handing it plans written by stronger models.789 The most common complaint is shallow or mistake-prone results on more complex work,101112 where several prefer Opus or GPT models.131415
By version sentiment of people who named that release, latest release first
Versions with at least 5 people. 16 people didn't name a version. Small samples swing a lot, so read the counts too. Compare all models →
Why people recommend it
Speed · 3
Many call it very fast, including compared with Anthropic models.
“Composer is actually really good, and fast.”
“I use it quite a bit and it works well and is extremely fast for the tasks it was built for”
Show 1 more
“it is very fast, especially compared to anthropic models, but the result it produces, especially with more difficult tasks is very shallow.”
Low cost · 3
Commenters say it is cheap enough for daily use under budgets, with a usable free tier.
“It is just a cheap, fast, surgical workhorse.”
“that I use a lot of the time when I have fairly straightforward tasks”
Show 1 more
“I find it really hard to justify the other models for the performance/cost I'm getting with Composer 2.5.”
Routine tasks · 3
Described as a dependable daily driver for small, focused, straightforward work.
“I do the bulk of my work with Composer 2.5. It’s quite good for “going forward” on smallish tasks”
“It's my daily driver, it's fast affordable and with a bit of guidance gets the job done.”
Show 1 more
“that I use a lot of the time when I have fairly straightforward tasks”
Implementation partner · 3
Used to implement plans written by larger models such as Opus.
“Composer 2.5 - which remains incredible and honestly nothing else compares for me”
“It is just a cheap, fast, surgical workhorse.”
Show 1 more
“their free tier of Composer is good enough”
Token efficiency · 3
Some say it uses fewer tokens while matching larger models on some tasks.
“Incredibly token efficient, fast, and gets the same quality as if you worked the whole thing with Fable/Opus/GPT.”
“Composer 2.5 is itself a fine-tune of K2.7, and it's apparently quite token efficient”
Show 1 more
“It's my daily driver, it's fast affordable and with a bit of guidance gets the job done.”
Code search · 1
One user finds it fast for code search, explanation and Q&A.
“Composer 2.5 is really fast in their harness at code search/explanation/Q&A tasks”
Caveats and dissent
Complex tasks · 3
Users report shallow or weaker results on more difficult tasks.
“it is very fast, especially compared to anthropic models, but the result it produces, especially with more difficult tasks is very shallow.”
“Incredibly token efficient, fast, and gets the same quality as if you worked the whole thing with Fable/Opus/GPT.”
Show 1 more
“Composer hands me a quickly put together college project that was started the night before it was due.”
Mistakes · 2
Some say they spend time cleaning up its errors and rate it below GPT models.
“Every time I use Composer 2.5 I have to spend a bunch of time cleaning up its mistakes.”
“Composer 2.5 is worse than both; I use it all day for simple stuff.”
Needs supervision · 2
Commenters say it needs guidance or can leave hidden problems unless watched.
“It's my daily driver, it's fast affordable and with a bit of guidance gets the job done.”
Benchmark doubts · 1
One commenter cites third-party benchmarks ranking it far below Cursor's own claims.
“Artificial Analysis' testing shows Composer 2.5 to be pretty far behind”
Often compared with
Who recommends it in their own words
Threads where it came up
- CursorBench 3.1 8 mentions
- SpaceX to buy Cursor AI coding agent operator Anysphere for $60B 6 mentions
- Cursor Introduces Composer 2.5 6 mentions
- Kimi K2.7-Code: open-source coding model with better token efficiency 3 mentions
- SpaceX and Cursor partnership. Right to acquire Cursor later this year 3 mentions
- Our decision on Cursor following its acquisition by SpaceX 2 mentions
Also written as Composer, Composer 2, Cursor Composer 2, Composer 1. Data: cursor-composer.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 48
“I do the bulk of my work with Composer 2.5. It’s quite good for “going forward” on smallish tasks”
“Composer 2.5 - which remains incredible and honestly nothing else compares for me”
“It is just a cheap, fast, surgical workhorse.”
“can be handled with Auto mode or Composer 2.5 (or Grok 4.6 High)”
“auto picks between Opus 5, GPT 5.6, Grok 4.6, and Composer 2.5”
“their free tier of Composer is good enough”
“Why would I need Composer unless I'm a poor hobbyist?”
“happily using Cursor + Composer 2.5/Opus”
“Built an awesome 450k line app with composer 2.5 and grok 4.5. I love it”
“I use Cursor and am happy with Composer :)”
“my Composer example speaks for itself”
“Composer 2.5 is itself a fine-tune of K2.7, and it's apparently quite token efficient”
“Like GLM, (now with 4.5) Grok, Composer.”
“Composer 2.5 was a lot better than I had expected”
“Composer 2.5 finetuned Kimi K2.5”
“Composer 2.5 is worse than both; I use it all day for simple stuff.”
“Every time I use Composer 2.5 I have to spend a bunch of time cleaning up its mistakes.”
“Kinda funny that their "cost-vs-performance" chart looks the same as the one for Composer 2.5”
“other models (GPT*, GEMINI*, Cursor - Composer 2.5 etc) as more like workhorse models”
“Incredibly token efficient, fast, and gets the same quality as if you worked the whole thing with Fable/Opus/GPT.”
“Composer 2.5 (brilliant and fast)”
“it is very fast, especially compared to anthropic models, but the result it produces, especially with more difficult tasks is very shallow.”
“my distaste for Composer and my preference for Opus”
“Composer hands me a quickly put together college project that was started the night before it was due.”
“I find it really hard to justify the other models for the performance/cost I'm getting with Composer 2.5.”
“I use for everything like trigger and monitor a 10 steps release process using composer , a very capable model”
“Artificial Analysis' testing shows Composer 2.5 to be pretty far behind”
“It's my daily driver, it's fast affordable and with a bit of guidance gets the job done.”
“I use Composer (since we use Cursor) or GPT 5.3-codex as my workhorse models”
“when i use Cursor i use 90% Composer 2.5 fast”
“composer 2.5 - mediocre with hand holding and super fast”
“Composer is actually really good, and fast.”
Can't remember the last time I used AI tab complete; back in the day indeed.
“that I use a lot of the time when I have fairly straightforward tasks”
“mirror what Cursor has done with Composer 2.5”
“Composer 2.5 is really fast in their harness at code search/explanation/Q&A tasks”
“mostly using Composer 2.5, but I freely change the model when the need arises”
“I use it quite a bit and it works well and is extremely fast for the tasks it was built for”
“I agree, Composer 2.5 is really good.”
“Composer 2.5 is really solid”
“Composer 2 performed differently on evals than Moonshot.ai's coding models”
“I use composer-2 daily for complex programming tasks.”
“They really can't produce much more than a function or a small module without shitting the bed”
“I've experienced Composer 2's performance and it leaves much to be desired as a daily driver for a knowledge worker.”
“I use composer-2 and sonnet 4.6 on a regular basis”
“I used to hate on Composer 2 but I'm coming around to it.”
“Composer-2 seems very useful in Cursor”
“Composer 2 is legitimately so impressive”