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

SWE-1.7 Reach Near GPT 5.5 and Opus Intelligence

cognition.com

Post reply on HN
Print view View original

71–80 of 151 posts

Re: SWE-1.7 Reach Near GPT 5.5 and Opus Intelligence

#71

Post by bluelightning2k »

pants2 wrote: post #19

Kinda funny that their "cost-vs-performance" chart looks the same as the one for Composer 2.5[1], except that it includes Composer 2.5 at a completely different spot. What are the chances that CursorBench ranks Cursor's model highest, and Cognition's bench ranks Cognition's model highest? Both are to be RL'd from Kimi as a base model, BTW. I'd posit that it's not deliberate deception, but for both companies their tra…

Good observation.

I actually started typing the same point that the chances are actually high because of train/eval overlap then realised you answered your own question with that same observation.

It is interesting though!

Perhaps in some way this means we should decide which eval set aligns best with our taste?

Back to the blog post. This is an excellent write up of an excellent technical achievement.

I have a lot of respect for the Cognition/Devin (always "Windsurf" to me) and Cursor teams.

I found it interesting - but justified - that they referred to themselves as a foundation lab rather than a dev tools company.

Re: SWE-1.7 Reach Near GPT 5.5 and Opus Intelligence

#72

Post by BeetleB »

llmslave wrote: post #33
wongarsu wrote: post #29

Earlier quoted context omitted.

At work I wouldn't want to use anything else. Compared to my salary a Claude subscription (or two) is cheap For hobby projects I've completely switched to DeepSeek v4 pro. I spend less than on a $10 Claude plan and am not subjected to quota limits (when I have time and motivation, the last thing I want is a 5 hour quota running out). And the difference in model performance is fine for those smaller projects, most of…

ive just got better things to do with my life than fuss with an inferior model. its like why hire a dumb employee over a smart one

I think you misspelled "I've got plenty of money".

Re: SWE-1.7 Reach Near GPT 5.5 and Opus Intelligence

#73

Post by sigbottle »

harmonic18374 wrote: post #2

A company whose first demo was completely fraudulent announces that its model beats GPT-5.5, on its own benchmark? I’m gonna wait a little before I trust this. This whole company seems to optimize for raising money and impressing VCs. Lying about their products, ignoring consumer market to target enterprise, bragging about how they work their employees like slaves, and writing these posts full of intimidating technic…

I highly respect many people at cognition but yeah that's put a sour taste in my mouth.

I want to work in the AI space on actual AI research, at any part of the stack. Even if I'm developing training infra - as long as people are advancing knowledge of what intelligence could be.

But it seems like either it's big labs or grifters, that's it, and even the big labs, at least publicly, seem very grifty at times. Not like I have the technical chops probably, but still.

Re: SWE-1.7 Reach Near GPT 5.5 and Opus Intelligence

#76

Post by anthonypasq »

lnenad wrote: post #65
Mitchem wrote: post #60

While I am skeptical of the results here, I am very excited for this new trend of making models faster. Running capable models at 1k TPS is more valuable for me than running better models at 30 TPS. I can only imagine the trend continues to move from "let's only make models smarter" to just incremental intelligence gains but with step improvements in speed.

Why? I'm personally on the opposite end. Less babysitting/higher quality means more time goes back to me/the user. 1000tps of bad code means you have to keep validating the output and circling back.

id rather iterate multiple times than wait 15 minutes to notice it made a mistake.

Re: SWE-1.7 Reach Near GPT 5.5 and Opus Intelligence

#78

Post by llmslave »

BeetleB wrote: post #72
llmslave wrote: post #33

Earlier quoted context omitted.

ive just got better things to do with my life than fuss with an inferior model. its like why hire a dumb employee over a smart one

I think you misspelled "I've got plenty of money".

200 bucks a month?

Re: SWE-1.7 Reach Near GPT 5.5 and Opus Intelligence

#79

Post by lnenad »

anthonypasq wrote: post #76
lnenad wrote: post #65

Earlier quoted context omitted.

Why? I'm personally on the opposite end. Less babysitting/higher quality means more time goes back to me/the user. 1000tps of bad code means you have to keep validating the output and circling back.

id rather iterate multiple times than wait 15 minutes to notice it made a mistake.

Again, my point is exactly the opposite. Higher quality implies a mistake isn't made in a significant % of cases.

Re: SWE-1.7 Reach Near GPT 5.5 and Opus Intelligence

#80

Post by unshavedyak »

lnenad wrote: post #65
Mitchem wrote: post #60

While I am skeptical of the results here, I am very excited for this new trend of making models faster. Running capable models at 1k TPS is more valuable for me than running better models at 30 TPS. I can only imagine the trend continues to move from "let's only make models smarter" to just incremental intelligence gains but with step improvements in speed.

Why? I'm personally on the opposite end. Less babysitting/higher quality means more time goes back to me/the user. 1000tps of bad code means you have to keep validating the output and circling back.

So i agree with you, but there's no SOTA model that i don't have to babysit. I'm not going to just throw a large pile of code in there unreviewed, and so what i want is faster iteration on code in logical, reviewable chunks. Ie just like i'd normally write myself; small, logical commits.

Faster iteration means i mentally checkout less and am more involved with the code being created.

My hope is that in the far far future, we can get LLMs so fast that i can work in my IDE like normal and the LLM will just be an extension of autocomplete. I can state a goal, rough out functions, code, etc, and it'll just work around me like a very fast pair programmer / autocomplete.

The chat interface is an intermediate step that frankly i hate. The faster it is the less i wait.

Now for vibe-slop i'm making on the side, yea i don't care about speed. But that's not something i'm employed to do or anything i truly care about. It's a different workflow entirely.

Post reply on HN