20 comments

  • paxys 19 minutes ago
    > Separate question, separate table. This is our standard latency harness (three short prompts, five reps, 400-token cap), not the build tasks. tok/s is output tokens over wall-clock, uniform for all.

    > so their tok/s is a ceiling, not a true decode rate. The clear read: the GPT-5.6 tiers are the snappiest models here on short prompts (Luna answers in about a second), Qwen is absurdly cheap and fast, and DeepSeek and GLM are the slowpokes

    You put in a lot of good work, and kudos for that, but man, reading paragraphs like these just puts me off of the entire piece.

    Like…how hard would it have been really to type these two sentences by hand, in your own natural voice?

    • fluidcruft 7 minutes ago
      Yeah, I can't figure out where this "voice" comes from and it is so impossible to get rid of. It is so grating.
      • Chu4eeno 5 minutes ago
        The problem is that everyone (well, too many) is now using the same "voice" when writing.

        So no matter how good or thoughtful the writing is it gets tiresome.

    • jakubmazanec 9 minutes ago
      > how hard would it have been really to type these two sentences by hand, in your own natural voice

      On the other hand, do we have to complain about every seemingly AI written text?

      • paxys 7 minutes ago
        Yes? AI generated text is explicitly disallowed on HN, so it’s not crazy to expect that a similar standard be used for linked content.
      • fluidcruft 6 minutes ago
        It is just horrible writing style. It doesn't particularly matter that AI wrote it.
  • ttoinou 39 minutes ago

       "This isn't objective." Correct, and we are not pretending it is. We are not handing down a scientific verdict. 
    
    
    Actually, you are doing rational investigation in a fuzzy probabilistic new/emergent space, with open sharing to the world. I don’t understand why people downplay themselves and put on a pedestal others supposedly serious sciences.
    • minimaxir 26 minutes ago
      It's a preemptive defense against methodology cynicism seen often on sites including but not limited to Hacker News. I've been guilty of including such defenses myself over the years because I've gotten annoyed with receiving such cynicism.

      Look at the top comment on their previous HN submission: https://news.ycombinator.com/item?id=48839886

    • sixhobbits 22 minutes ago
      because if you don’t put this disclaimer the top comment is always "Acthually this isn't real science because you didn't publish your P value" so you can't win.

      also the article itself is clearly LLM generated though

    • boondongle 17 minutes ago
      Ultimately advocates exist for models and there are incredible financial incentives for some to be advocates, so authors are guaranteed someone being mad if their horse doesn't perform well.

      Given that type of reaction is inevitable, it just saves the conversation.

    • jakevoytko 18 minutes ago
      Because for its entire existence, the top HN comment on articles is typically a contrarian take or pointing out flaws. This goes double for a study, where people just hunt for some aspect of the methodology they dislike. If you don't address the flaws, then it looks like you never considered them, and the top comment will say that your entire methodology is suspect. It's super predictable to the point that you can harness this kind of reaction to get stuff on the frontpage if you really want to.
    • adammarples 29 minutes ago
      Because serious science is hard and valuable for its rigour, and shouldn't be compared with just poking at data to see what happens
      • chris_money202 26 minutes ago
        Don't be fooled, there is politics, opinions, and less rigor in science as well.
  • platinumrad 9 minutes ago
    Maybe I'm a control freak, but asking agents to one-shot random apps is nothing like how I actually use AI in software engineering.
    • bhu8 7 minutes ago
      Absolutely yes, but that's how you become twitter/X famous
  • thebigspacefuck 29 minutes ago
    (LM)Arena is basically this. IMO it’s the best benchmark that avoids benchmaxxing

    Agent: https://arena.ai/leaderboard/agent

    Web dev: https://arena.ai/leaderboard/code/webdev

    Currently Fable and 5.6 are neck and neck on web dev which is basically the same finding as this.

    • Chu4eeno 0 minutes ago
      There's a ton of arenamaxxing going on (especially from facebook), though I don't disagree that it's one of the better actual benchmarks.

      Always fun to ask them to recreate classic demoscene effects (sadly they're still pretty bad at generating music, though at least claude seems to create decent synths).

      I keep trying to get them to recreate the fluid+particle stuff from Agenda Circling Forth etc., but even giving them the blog posts describing the implementation (and screenshots) they're still pretty bad.

  • sgk284 43 minutes ago
    Similarly, we updated our model arena (52 apps each built by 26 models) to have GPT 5.6 Sol, Terra, and Luna today:

    https://arena.logic.inc/

    It's really interesting to see the Sol/Terra/Luna apps side-by-side.

    I need to add these stats somewhere in the UI, but one interesting take away: Terra took 1/2 as much wall-clock time as Sol, but Luna took more wall-clock time than Sol (by about 23%). It's still much much cheaper, but it seems like Terra is likely a more optimal time/cost balance for most use cases.

    The Terra quality is usually nearly as good as Sol, but much faster and cheaper. I do appreciate Sol's design sensibilities (see, for example, the audio sequencer). It's the first model in a while that is clearly distinct on that front. They'd all converged to very similar visuals for a while.

    • vitorsr 18 minutes ago
      What caught my eye was:

                  Model  Lines of Code  File Size  Gzip Size 
            GPT-5.6 Sol          1,264    35.5 KB    10.0 KB 
          GPT-5.6 Terra            827    20.0 KB     6.7 KB
      • sgk284 8 minutes ago
        Yea, that's an interesting result as well. The Terra apps don't feel 35% less feature-rich. So it seems quite token efficient.
  • ricardobeat 17 minutes ago
    Obviously AI-written, but I'm confused with the results: Muse Spark has the best Rubik's cube by far, the only one properly animating, yet it gets a 2/5

    (edit: seems to be an issue with inline videos)

  • master_crab 4 minutes ago
    A lot of these are visual-heavy tests that often require first person sight to confirm results. Considering GLM isn’t multimodal, that might explain why it did better on the calculator question and not much else.
  • rbehrends 24 minutes ago
    My concern with most of these visual benchmarks, popular as they are, is that they are likely more indicative of knowledge (i.e. how comprehensive the training data is and how well it can be retrieved from the model) than of reasoning ability. I don't see in particular how a model would construct a CoT that mapped somehow to a representation of the cube geometry and its animations in latent space without a large chunk of that being pre-existing information.
  • smusamashah 36 minutes ago
    "One honest caveat", "no glitches, no color changes" good tests and I read it to the end but I wish it was written by a human.
  • ianm218 41 minutes ago
    This does seem to validate the critique that models like GLM are benchmaxxed and not as close to the frontier as you’d think based on their numbers.
  • esafak 1 minute ago
    Could you make the tables sortable?
  • orliesaurus 13 minutes ago
    Missing the exact prompts - would love to replicate...but also curious how you prompted these: they could be a big reason why some models failed completely at rendering SVGs (ie. GLM 5.2)
  • orliesaurus 16 minutes ago
    Really nice breakdown, surprised by the results - especially the fact that OSS models were so behind on most task... (lol at the SVG of the moon without any sign of life by GLM-5.2)
  • dinkleberg 18 minutes ago
    Is this how I learn that Bezos now has a beard? Interesting that it is a detail that all of the models chose to include (unless that was in the prompt and just not put in the post).
  • kibae 38 minutes ago
    The cost seems to be using the wrong symbol: ¢ vs $
    • delichon 28 minutes ago
      Nope, they're that cheap. E.g. Grok 4.5 is $.02 to $.06 per million tokens. A 400 token reply costs ~.002¢

      https://www.tryai.dev/models/grok-4.5

      • il 24 minutes ago
        Grok 4.5 is $2/$6 there's no model anywhere close to that cheap
        • delichon 19 minutes ago
          The numbers come from the tryai.dev link:

            How much does Grok 4.5 cost on TryAI?
            Grok 4.5 is Input: $0.02 / 1M tokens, Output: $0.06 / 1M tokens. There is no subscription — you pay only for what you use.
          • kibae 13 minutes ago
            Those numbers are incorrect unless they got a deal that's 100x cheaper than API pricing which is unlikely. They updated the original post with the correct costs.
          • ricardobeat 13 minutes ago
            Slop pricing pages? https://www.tryai.dev/models/claude-fable-5 says Fable costs $0.1/$0.5. Can't wait to use it at those prices!
  • CompoundEyes 20 minutes ago
    It’s interesting how all the model names and versions are like SKUS taking up space on a display shelf. I look forward to whatever Sagittarius A* does!
  • joehabeebs 53 minutes ago
    Interesting tests being done but I can't help but think it limits testing innovation in some way given that the requested apps are essentially all clones of others
    • christophilus 49 minutes ago
      I hear this take a lot, but every app I’ve ever built was like 80% similar to every other app out there. The unique/ creative part of an app is not the bulk of it, and LLMs have been pretty good at helping me explore the 20%, too.
      • billyp-rva 37 minutes ago
        Calculator / Rubik's cube / game of life apps should be very close to 100% identical, right? I don't see the point of asking an AI for one of these when there are dozens (hundreds?) of repos that all have exactly what you want.
  • sangupta 21 minutes ago
    Sign-in via Google is broken - it redirects back to localhost from Supabase :)
  • throw310822 22 minutes ago
    "Elon and Bezos watch a Blue Origin landing" svgs are super cute, and incredibly like children's drawings. They also nail Bezos' features pretty well.
  • CharlesW 34 minutes ago
    > We generated a big pile of artifacts, we are publishing all of them, and you can form your own opinion.

    My opinion is that two gimmicky "one-shot prompting shootout" marketing pieces in two days smells like desperation. I'm not sure you understand what a turnoff this is for potential customers.