20 comments

  • bob1029 a few seconds ago

    [delayed]

  • weli 21 minutes ago

    Anecdotal, opinion:

    Gpt is really good in vision stuff, or at least their MoE seems to be really cohesive. From my experience Claude models can be really good at language but the moment they need to look at a picture and decide why the design is not good what parts need improvement it degrades a lot. My easiest benchmark is giving them a screenshot of a feature in my app and tell it "identify non-normative UI blocks and improve readability and consistency". Sol does a great job at re-structuring the page into composable units that build upon each other and the general looks and feels of the app. Claude tends to over-focus one one part while completely forgetting about the rest or the cohesion as a whole.

      velcrovan 6 minutes ago

      Assessing the subjective quality of a thing is in my experience one of the worst ways to use any LLM.

        rib3ye 2 minutes ago

        anthropic frontend-design skill does a great job with it.

      DaiPlusPlus 4 minutes ago

      What is a "non-normative UI block"?

        lelandfe a minute ago

        areas that look weird

  • evrimoztamur 9 minutes ago

    Penny sample shown looks like failed EXIF orientation registered by the model/harness. The coins are correctly marked, it's rotated 90 degrees.

  • kzrdude 12 minutes ago

    In the third vision bench result, Sol is 100% correct but the expected has 1 error. Seems like an oversight.

    In the next bench, Sol looks like it’s correct again but the bboxes are rotated 90 degrees for some reason.

  • sscaryterry 17 minutes ago

    My anecdotal evidence says its still as blind as any other model, it has no taste, no attention to any sort of detail.

      howdareme 15 minutes ago

      How can a vision model have taste?

        sarreph 11 minutes ago

        If you're doing any kind of inference that is multi-modal and non-factual, opinions and biases will affect any kind of assessment of a visual that you provide to a model.

        For example, a UI / UX professional being asked to appraise a website screenshot may determine that the image in question has "desirable" traits which are inherently not deterministically measurable. Such as, if the interface elements have strong information hierarchy, or if they are deemed to be "fashionable" with current UI trends.

          DaiPlusPlus 2 minutes ago

          > if the interface elements have strong information hierarchy

          ...but that's an example of a UX/usability matter that can be assessed objectively and non-subjectively.

        sscaryterry 11 minutes ago

        Replace taste with consistent if that helps you. Can it follow a design system...

          yreg 6 minutes ago

          As a design system engineer I usually have to fight against the taste of the designers. (And I consider it natural.)

          But, if you have a proper well documented design system and you tell the LLM to use the DS and to avoid styling hacks they can generally do it. Even the dumber ones than Sol 5.6.

          Of course only if the design is achievable in the design system.

            sscaryterry 4 minutes ago

            This is not my experience at all.

          velcrovan 5 minutes ago

          So, formulaic output…the opposite of taste

            sscaryterry 3 minutes ago

            Not really. Compliance with the letter of the law doesn't mean the intent is complied with.

  • Razengan 16 minutes ago

    For 2 weeks I've been trying to get Codex to "outpaint" a wonderful image it generated as placeholder art for a level background.

    After I increased the game's resolution, I asked it to increase the image's size while keeping the same scale and existing content, and gosh, it constantly keeps getting something wrong no matter what I tell it, even on Sol Max with the $100 Pro subscription.

    An average pixel-artist could have recreated the image and more within 2-3 days.

      sscaryterry a minute ago

      > it constantly keeps getting something wrong no matter what I tell it

      This 100%

      thatcat 12 minutes ago

      did you try segmenting it first?