Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Composer: Building a fast frontier model with RL

cursor.com

Post reply on HN
Print view View original

121–130 of 169 posts

Re: Composer: Building a fast frontier model with RL

#121

Post by pbowyer »

netcraft wrote: post #107

I love cursor, the tab completion and agent mode. But I really dislike vscode after using intellij for so many years. I really wish the underlying editor was better, or I could get cursor features in intellij instead. The editing of the files is mostly fine, but its everything else around it that a full IDE provides thats just so much better. Right now its intellij + claude code for me, and its fine, but I wish I cou…

Intellij's tab-complete is coming along; it's hit and miss if it will work but for similar edits I'm finding it picks up the pattern quickly and I can tab - tab - tab to make them happen.

Still not up to Cursor standards though :)

Re: Composer: Building a fast frontier model with RL

#122

Post by coder543 »

srush wrote: post #16

Hi everyone, I am an ML researcher at Cursor, and worked on this project. Would love to hear any feedback you may have on the model, and can answer question about the blog post.

Impressive systems write-up. A question: if Composer is an RL finetune on an open model, why keep weights closed? The edge from a slightly better checkpoint erodes quickly in this market, it's not a durable advantage. Composer protects Cursor's margins from being squeezed by the big AI labs, but that is true whether the weights are open or closed, and I think Cursor would have more lasting benefit by generating developer goodwill than from a narrow, short-lived advantage. But, that's just my opinion. I personally find it hard to get excited about yet-another proprietary model. GPT-5 and Sonnet 4.5 are around when I need one of those, but I think the future is open.

Re: Composer: Building a fast frontier model with RL

#123

Post by NaomiLehman »

jasonjmcghee wrote: post #11

Maybe I'm an outlier but Sonnet 4.5 quality is about as low as I'm willing to go. It's generation speed is not the problem or the time sink. It's wrestling with it to get the right output. --- And just to clarify as maybe I misunderstood again but people are comparing cursor to Claude Code and codex etc here- isn't this whole article all cursor just using different models?

gpt-5-high is as low as i can go :]

Re: Composer: Building a fast frontier model with RL

#124

Post by vidarh »

lubujackson wrote: post #91
vidarh wrote: post #31

Earlier quoted context omitted.

I feel like that's like having a lead in producing better buggy whips. I run Claude Code in the background near constantly for a variety of projects, with --dangerously-skip-permissions, and review progress periodically. Tabbing is only relevant when it's totally failing to make progress and I have to manually intervene, and that to me is a failure scenario that is happening less and less often.

This is just a completely different use of LLMs and has little to do with working at a real business with a live site and users. Cursor is great when you want to gain understanding of an issue quickly, or resolve something clear and specific quickly. I'm not against YOLO vibe coding, but being against tab completion is just insane to me. At the end of the day, LLMs help you achieve goals quicker. You still need to kn…

Some of these projects are at a "real business with a live site and users". Two of the current ones are.

And it's not remotely "YOLO vibe coding". All the code gets reviewed, and tested thoroughly, and they are worked to specs, and gated by test suites.

What I don't do is babysit the LLM until it's code passes both the test suite and automated review stages, because it's a waste of time.

Others of these projects are research tasks. While I wrote this comment, Claude unilaterally fixed a number of bugs in a compiler.

Re: Composer: Building a fast frontier model with RL

#125

Post by vidarh »

camdenreslink wrote: post #105
vidarh wrote: post #31

Earlier quoted context omitted.

I feel like that's like having a lead in producing better buggy whips. I run Claude Code in the background near constantly for a variety of projects, with --dangerously-skip-permissions, and review progress periodically. Tabbing is only relevant when it's totally failing to make progress and I have to manually intervene, and that to me is a failure scenario that is happening less and less often.

What are you building with this workflow? Is it an application live in production with users? It is such a foreign way of working to me.

A compiler (hobby project). A web application server (tooling for my consultancy). An agentic framework to part-automate end-to-end development of a large web app (customer project). An analytics platform to analyze infrastructure maturity (customer project).

Usually I'll have several Claude Code sessions running in parallel on different projects, and when one of them stops I will review the code for that project and start it again - either moving forwards or re-doing things that have issues.

Re: Composer: Building a fast frontier model with RL

#126

Post by criemen »

cwyers wrote: post #90

The lack of transparency here is wild. They aggregate the scores of the models they test against, which obscures the performance. They only release results on their own internal benchmark that they won't release. They talk about RL training but they don't discuss anything else about how the model was trained, including if they did their own pre-training or fine-tuned an existing model. I'm skeptical of basically ever…

I understand where you're coming from, and I'd love to have learned about pre-training vs. off-the-shelf base model too. But

> their own internal benchmark that they won't release

If they'd release their internal benchmark suite, it'd make it into the training set of about every LLM, which from a strictly scientific standpoint, invalidates all conclusions drawn from that benchmark from then on. On the other hand, not releasing the benchmark means they could've hand-picked the datapoints to favor them. It's a problem that can't be resolved unfortunately.

Re: Composer: Building a fast frontier model with RL

#127

Post by ripped_britches »

srush wrote: post #16

Hi everyone, I am an ML researcher at Cursor, and worked on this project. Would love to hear any feedback you may have on the model, and can answer question about the blog post.

Amazing work! The UX is great.

GPT-5-codex does more research before tackling a task, that is the biggest weakness for me not using Composer yet.

Could you provide any color on whether ACP (from zed) will be supported?

Re: Composer: Building a fast frontier model with RL

#130

Post by dlojudice »

srush wrote: post #16

Hi everyone, I am an ML researcher at Cursor, and worked on this project. Would love to hear any feedback you may have on the model, and can answer question about the blog post.

Congratulations on your work. I spent the day working with a mix of the Composer/Sonnet 4.5/Gemini 2.5 Pro models. In terms of quality, the Composer seems to perform well compared to the others. I have no complaints so far. I'm still using Claude for planning/starting a task, but the Composer performed very well in execution. What I've really enjoyed is the speed. I had already tested other fast models, but with poor quality. Composer is the first one that combines speed and quality, and the experience has been very enjoyable to work with.
Post reply on HN