@IAlwaysEatCats @addiswag1 Yeah butcher and hunter are both strong as hell
English
Jack
833 posts
@addiswag1 I sent three butchers and 1 hunter into the future and they killed hitler
English
How life feels when you stop making mewgenics teams based on what’s “Meta” and just do the classes you find fun to play
English
Why GPT-5.5 is lower than Claude?
The answer is simple: Code Arena currently only supports frontend/web development tasks, where GPT-5.5 is weakest. Full-stack app development and GitHub integration will land in a couple months.
Next time we'll be clearer that this leaderboard shows React/FE only, until we ship full-stack apps etc. Thanks for the feedback!
Arena.ai@arena
GPT-5.5 by @OpenAI is now live in the Arena, landing across multiple leaderboards. Here’s how it ranks by modality: - Code Arena (agentic web dev): #9, a strong +50pt jump over GPT-5.4 - Document Arena (analysis & long-content reasoning): #6, on par with Sonnet 4.6 - Text Arena: #7, Math #3, Instruction Following: #8 - Expert Arena: #5 - Search Arena: #2 - Vision Arena: #5 Strong, well-rounded performance, especially in Code (+50 pts vs GPT-5.4). Congrats to @OpenAI on the release. Full category breakdowns by modality in the thread.
English
Wow what an extreme thing to say for the sake of humor. I wonder what he posted to subvert our expectations
QT Patch@WITT_SZN7
Lane Thomas walkoff and I’ll post my fully erect shaft and balls
English
@synthwavedd Which story do you think they’re gonna run with this time? They’re A/B testing the plans, or this was just an unintended bug?
English
@yuzu_4ever I’d also wager Tim’s poll is larger in scope, given that Twitter is accessible to anyone with internet. I think the US is among the generally higher trust societies that would skew more blue, which is not the case globally.
English
@0ranguchad sorry i wasn't being clear. i didn't mean larger in the sense that it had more votes than the twitter poll.
English
i would rather die than live in a world only composed of red button pressers.
Tim Urban@waitbutwhy
Everyone in the world has to take a private vote by pressing a red or blue button. If more than 50% of people press the blue button, everyone survives. If less than 50% of people press the blue button, only people who pressed the red button survive. Which button would you press?
English
@yuzu_4ever Tim’s poll is way larger? 98k responses versus 2.6k. Also, again, there’s no actual threat of death being tested in this poll.
English
@0ranguchad idk in larger studies the gap is much wider. twitter skews a certain way. x.com/davidshor/stat…
David Shor@davidshor
We asked this to a large sample of nationally representative Americans - blue wins by a 3:1 margin!
English
The IQ mogging continues on PencilPuzzleBench
Lisan al Gaib@scaling01
two facts - Opus 4.7 is a decent upgrade. if it's worse for you it's a skill issue - GPT-5.5 will absolutely IQ mog Opus 4.7
English
@yuzu_4ever Which is not to say that I think everyone who’s picking blue is being dishonest, but that the majority of people, when confronting the actual finality of death, will choose to live. I’d predict ~80-90% red in the final poll.
English
@yuzu_4ever That Tim’s poll is barely a win for blue when considering the sample size, lack of stakes, and the population being polled should be a very strong indication that red would win IRL. I’d like to say I’d still pick blue, but I think the honest answer is I wouldn’t self-sacrifice.
English
@yuzu_4ever Pick it. I would mourn the loss of idealists whom I believe to have been misled, and I definitely think framing the question differently would produce different results (eg the blender hypothetical). Still, I think self preservation is the dominant factor for most humans.
English
@yuzu_4ever Ultimately, I’m red because I believe that when *real stakes* are offered, the majority of people will choose self preservation. The finality of oblivion is far more threatening to confront than a Twitter poll. Logic dictates that red is the direct solution, so everyone should
English
@redtachyon Claude’s sycophancy is not new. It’s baked into the model.
English
Genuine question - how do people get this delusional? Is it some new 4o-ish sycophancy effect from Claude?
Anthropic and OpenAI are competitive with each other in SWE/coding. IMO Codex is far better than Claude, but I know smart people who think otherwise, and that's fine. There's no objective winner, they're just comparable products.
At the same time, OpenAI also has some bets in image/video generation, world models, who knows what else.
I see absolutely no moat that Anthropic has over OpenAI, there's no world in which they "win", EXCEPT for a fast-takeoff ASI that they potentially create before anyone else.
(looking forward to being muted - I tend to not block people though)
Zhu Liang@paradite_
I’m muting people on X who don’t understand how far ahead Anthropic is. It’s hurting my brain so much that I had to do this. If you think OpenAI is in anyway better than Anthropic, please just block me to save both us some trouble.
English
so fun to see the reception to 5.5!
there is almost nothing that feels more gratifying to me than builders saying they find our tools useful.
English
