I'm impressed that Luna got as far as it did on max reasoning. Don't get me wrong, that's one weird looking ass, but even the top GPT model a year ago probably would have done worse based on my experience (with asking it to create meshes in general, not modeling asses... yeah, that's the ticket). Then again, I'm also not, because I've found it to be a great deal at its relatively low price point when the reasoning is set to high.
still awkward when a page full of thong-covered asses pops up on my extra monitor in my open-seating arrangement office that im required to attend 4 times a week (no of course im not salty or anything)
i almost wrote a extension to block mentions of pelican bench because i find those so utterly useless. now, Assbench, on the other hand, i could get behind
Finally, I was getting tired of seeing pelicans on a bike.
Up next, Pelican ass-bench!
6 months from now we'll be discussing if model training is assmaxxing.
Why is this flagged? Isnt the readership adults? (Do children want to read these dry articles on tech?)
I'm impressed that Luna got as far as it did on max reasoning. Don't get me wrong, that's one weird looking ass, but even the top GPT model a year ago probably would have done worse based on my experience (with asking it to create meshes in general, not modeling asses... yeah, that's the ticket). Then again, I'm also not, because I've found it to be a great deal at its relatively low price point when the reasoning is set to high.
The idea that they're going to benchmaxx this in the future amuses me
Folks, we've reached the top.
It's about bottoms though...?
ASStra for the win!
how come opus 5.5 did so well? I'm surprised
Na, not well at all. Now Fable, fable like'em thick.
Because it is ass.
Better data set?
https://www.assbench.com/PROMPT.md sent back an error.
Error code: 404 Not Found
The prompt that created these images doesn't mention that they want a Caucasian skin color. Huh. Racist LLMs?
Makes sense. Most look like pancake butts or shitty implants.
It’s only a sample of one per model
Biased data?
The most important of benchmarks.
Glad my work blocked this. But can we get a NSFW tag? If this is NSFW?
If I were you I wouldn't rush to open a website that clearly has "ass" in the title
It’s fine. You’ll be fine
I go to church every week. I try not to swear.
The site is fine, and it's funny to my inner 13-year-old boy.
Next, HumanBench! Draw an entire, picture-perfect human.
nsfw?
but hilarious nonetheless
It's ok they're wearing thongs it's not nudity.
still awkward when a page full of thong-covered asses pops up on my extra monitor in my open-seating arrangement office that im required to attend 4 times a week (no of course im not salty or anything)
7 replies →
presumably that's why it was flagged? is NSFW against HN's policy?
Agree 2x.
But can we make thicc pelicans?
haha, doing the important work!
Origami Ass
Dreamed up by Claude "Haiku"
Unironically
折り紙臀 (origami den)
ハイクの作品 (Haiku no sakuhin)
皮肉のない (hiniku-no-nai)
the eval we need
the eval we don't deserve
i almost wrote a extension to block mentions of pelican bench because i find those so utterly useless. now, Assbench, on the other hand, i could get behind
Opus 5.5 is SI
We need Banana Bench next.
Now put them on a bike.
[flagged]
Yeah, the HR people will freak out. Any excuse for those guys to come down like a hammer.
All right won’t post but this is awesome
I wouldn't, unless wanting to get fired.
[flagged]