Hacker News Viewer

LLM Ass Bench

by fragmede on 9/22/2026, 8:34:19 PM

https://www.assbench.com/

Comments

by: ravenstine

I&#x27;m impressed that Luna got as far as it did on max reasoning. Don&#x27;t get me wrong, that&#x27;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... <i>yeah, that&#x27;s the ticket</i>). Then again, I&#x27;m also not, because I&#x27;ve found it to be a great deal at its relatively low price point when the reasoning is set to high.

9/22/2026, 10:17:21 PM


by: drywater2

Finally, I was getting tired of seeing pelicans on a bike.

9/22/2026, 9:14:44 PM


by: sethkim

Folks, we&#x27;ve reached the top.

9/22/2026, 10:09:18 PM


by: sergiotapia

But can we make thicc pelicans?

9/22/2026, 10:27:25 PM


by: athrowaway3z

6 months from now we&#x27;ll be discussing if model training is assmaxxing.

9/22/2026, 10:15:45 PM


by: TutleCpt

The prompt that created these images doesn&#x27;t mention that they want a Caucasian skin color. Huh. Racist LLMs?

9/22/2026, 9:56:47 PM


by: ramon156

how come opus 5.5 did so well? I&#x27;m surprised

9/22/2026, 9:49:09 PM


by: variety8675

The idea that they&#x27;re going to benchmaxx this in the future amuses me

9/22/2026, 8:44:27 PM


by: spottedmarley

ASStra for the win!

9/22/2026, 9:05:40 PM


by: danbrooks

Glad my work blocked this. But can we get a NSFW tag? If this is NSFW?

9/22/2026, 10:05:41 PM


by: VCFundedGenYer

The amount of blatant slop I&#x27;ve flagged in the past few hours is insane. HN mods need to do something.

9/22/2026, 10:21:30 PM


by: visiondude

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

9/22/2026, 10:07:57 PM


by: Avicebron

&quot;The economy was in a state of deep neglect, a great dust bowl had ravaged food supplies. And the number one movie in the country was called Ass. And that&#x27;s all it was, for 90 minutes.&quot;

9/22/2026, 10:14:14 PM


by: Ancalagon

nsfw?<p>but hilarious nonetheless

9/22/2026, 9:08:07 PM


by: winddude

haha, doing the important work!

9/22/2026, 9:48:08 PM


by: fragmede

The most important of benchmarks.

9/22/2026, 8:35:05 PM


by: Rapzid

We need Banana Bench next.

9/22/2026, 9:58:30 PM


by: randomblock1

Now put them on a bike.

9/22/2026, 10:03:45 PM


by: threethirtytwo

I really want to post this on my work slack cuz it’s hilarious. Should be fine right? This should be sfw technically imo but I want consensus here. I don’t want some crazy Mia or Lauren girl to report this to HR (they don’t actually care but they want to fuck my shit up cuz they consider me to be an “op”). Just want to confirm.<p>Also I’m just coming out of a surgery and I’m sitting here waiting for my ride and looking at HN, so my judgement is impaired please tell me if I should&#x2F;can post.

9/22/2026, 10:10:39 PM