Really happy for Adam and the rest of the crew - they made a great product for a prior era, and hopefully this is a soft landing for them given their revenue stream dried up.
Tailwind at this point is a coding language and a brand with no direct product. It's a great acquisition when you want mind share and developer love, and are OK with there being no revenue involved. Shopify is a solid match for that.
I use gpt image 2 very heavily for my current project (ai UI design tool). The biggest improvement I'm seeing with this is in speed. I've generated around 50k images with gpt-image-2 via api, the the average latency has held at around 104s.
It's wild how much of a difference this is - images are coming in at around 35-40s. Very noticable, and makes a difference when you're iterating quickly: https://jjcm.org/gpt-image-2.5-speed.mp4
Overall it used the reference images I gave it a bit better than gpt-image-2. I noticed 2 had issues getting the blue button just right. 2.5 nailed it.
Did very well modifying the pose while keeping the appearance of Glenn Powell. gpt-image-2 had a lot of the "fried" look for some of his skin in prior designs I did for johnpolitics.com
Dark mode sites surfaced the fried look quite a bit in prior models, but this definitely looks better on that front. One thing that looks perhaps worse though is the microglyphs - note the teal lines to the bottom right of the ramen, they're kinda blurry / not straight.
Overall fixed some of the main issues / gripes I had with gpt-image-2
The 'cyberpunk' one is incredibly bland and uninspired. That may be the goal of the political site, as a form of satire.
But reducing a style with a wealth of cultural material to draw on (literature, fashion, cinema, gaming, industrial design) to 'dark with turquoise/pink highlights, cut corners on rectangles, some random diagonal lines, some token Japanese characters and a lazy neon-style logo' feels like a miss.
If this was from a human designer I was working with, I'd be having a serious conversation about effort.
It's both. I have several models loaded into diffui. Which one each prompt uses is determined based on user preference over time. Whichever image is currently at the top of any image node stack marks a win for the model that generated it, I assign each model an ELO score based on that, and I bias the chance each model is selected based on their ELO score. Right now gpt-image-2 is better than my own model, and it services around ~96% of the requests in diffui.
I'll also be adding in microsoft's mai-image-2.6 soon, but I need to update my SOC2 to add MS as a provider before I turn that on for other users. The full list of models in rotation is here: https://image.non.io/d53a9760-8b74-4386-b032-d59da2cd5319.we...
Oh, you're the dev of diffui? It's completely off-topic, but I've notice that RMB -> Download image will download an .png but it's actually a .webp and it can cause issues (e.g. file explorer doesn't show thumbnail correctly).
IMO one of the biggest losses of the OpenAI/Cursor breakup will be the loss of OAI models on CursorBench [1]. Their bench has always been one that most-fit my mental model of how good each of these models are. I find AA’s Intelligence index to often be out of alignment with my own subjective evals.
Note the flowing svg lines, and how accurately it recreated them. Here's Opus 5 for comparison - you can really see how while Astra really recreated the flow that was in the original design, opus only got the general vibe: https://image.non.io/dfe13de0-4487-431f-8b69-544ff3030dac.we...
One thing I will say is you are paying for quality. That site build cost $24 - extremely non-trivial for a simple frontend.
A bespoke design like that a year ago would have cost $200-2000+ between design and development. Hell, it probably still does unless the person wanting the design is already a developer who knows how to prompt.
Everything costs more now than it did a year ago... except for THIS, and we are still complaining that a 90-99% reduction in cost is STILL too expensive. And a 50-75% reduction in time is STILL too long.
We used to have to wait for weeks for a design like that when I worked at a consultancy, and that is a week of salary. For the design, then it got handed off to a front end developer to slice it and get built so the back end developer can hook it up to a CRM. We are talking a month turn around with design, revisions, development, testing, and bug fixing.
It can now be done in a couple of hours for less than a single hour's cost. If it were 10x slower and 10x more expensive, it would STILL be "good deal".
Wdym no room for error or experimation? Changes are even cheaper, and in my experience they are faster and way cheaper with models than with designers.
I would say opus was in some ways more accurate, but missed the higher level curvature feel of the site that astra picked up on.
It sounds completely trivial and likely I'm wrong here, but could it be that opus saw the reference image squished? That might explain the sharper horizontal curvature
Opus 5 is quite good, and has been the SOTA for this test (by my own subjective comparison) for the last few months. One thing it's always struggled with though is recreating smooth SVG curves/cutouts.
Another way to think about it is you could prompt Astra to put the building back in. You couldn't prompt Opus 5 to get the correct curvature of the line / cutout. That part has always been a huge struggle for models.
Ontological discussions aside, I’ll share out my recipe for one of my favorite variations of a Caesar that embraces its origins. It’s one my friends usually request I make/bring.
It’s composed of four main parts, made in this order: the the meat, the croutons, the dressing, and the leaf.
For the meat we’re making a carne asada. You’ll want a flank or a skirt steak, or anything thinner with good marbling. For the marinade use olive oil as the base. I generally do around 2 cups of olive oil for a larger serving. Don’t sweat the exact amounts for a marinade. Add around 10% Apple cider vinegar, 10% red wine vinegar, and 10% lime juice. Use garlic, cumin, black pepper, salt, smoked chipotle peppers (you can also find them in an adobo sauce which works great - skip the sugar if you use the adobo), two spoonfuls of sugar.
Blend all of it together heavily, then pour it over the steak. I like a smoothie blender for this as they tend to work better for these small batches. IMPORTANT: don’t wash the blender, we’ll use the remains each time to influence the subsequent thing. Marinate the steak for at least 2 hours.
Next we’ll make the croutons. Store bought and homemade croutons are very different. For the marinade add a full clove of garlic into our used blender, black pepper, salt, two spoonfuls of nutritional yeast, a thumb sized portion of Parmesan, and rosemary. Add oil so it fills in the gaps in the blender. The goal is a paste, not a liquid. Blend it up. Cut up the bread into slices, then tear it into chunks. The tearing creates better texture. Put the paste in a bowl, smear it all around, dump the bread into, and toss it so it’s coating all the bread. Put it in the oven on a low broil (grill for those in Australia). Careful as it can burn easily. Check it every minute til browns, toss it, then put it back in. Repeat til the outer shell is dry but the inner part is still squishy. They shouldn’t be wet. Remove and chill for 10min before adding them.
For the dressing, half olive oil, 15% lime juice, 15% white wine vinegar, 10% pepitas, 10% Parmesan. This is your base. Add an egg yolk, black pepper, a squirt of mustard. Blend it up, taste it, adjust as needed. I skip the anchovy for the Mexican vibe this salad has as I don’t think it resonates well, but by all means go for it.
Grill the steak high heat.
For the leaf any salad leaf will do, but add radish slices as well.
Toss the leaf and the dressing, top with crouts, steak, pepitas, and Parmesan. Enjoy. Should take you 60-90min of prep.
I've actually had a lot of success with something similar for my startup. I'm a first time founder, and there are so many unknowns when you're getting started. I've been using a Hermes agent to act as my "boss", giving me three tasks to do each day / creating a memory of everything with the business.
It's really, really helpful. So many things I would have completely missed if it weren't for this. I don't think you need this synthetic org setup to get exec help at a small level. It really helped me with fundraising, incorporation, compliance, and just keeping me on track.
Based on what? How do you judge if those are the right things to work on? As a founder, your number 1 priority is essentially to manage that the company works towards the right thing, but it sounds like to me you're outsourcing that part to a LLM? How do you know what the "right track" is when the track is just what the LLM outlined for you?
> So many things I would have completely missed if it weren't for this.
This is the "unknown unknowns" problem. It sounds like the agent is a reasonably good source of obvious-in-hindsight problems that a experienced founder would have experienced, but a new founder wouldn't easily guess at.
> judge if these are the right things
This problem exists no matter where you get the things from.
I do this with a Gemini instance that is connected to my email and calendar and task list. Every morning I get an email with an agenda. A lot isn't explicitly taken from my task list, but is instead inferred from the contents of emails.
As a mundane example, I had an email from my wife about a kid's swimming party she'd RSVP'd to.
My agenda that morning reminded me to pack, inter alia, sunblock and towels for my new car's seats. So it inferred sunblock from the party mention, and towels from the party mention alongside emails talking about my new car (and some "knowledge" that people take more care to protect new cars vs old cars).
This is a fantastic idea and I'd like to add my voice to the choir asking you for more information :) Don't let us rebuild everything from scratch if we don't have to
Would love to see how do you setup the agents, I'm not at the level of director but I think it would be good too for staff engineer to manage the office politics
Bartosz Ciechanowski really gave us a look at what the future of the web would look like. It seems like these kinda pages are the norm now with AI assisted dev, which I absolutely love. Fully interactive pages are so much more intuitive than static ones of the past. He has my thanks for ushering this in as a standard.
AI definitely lowers the cost of building these, but I think Ciechanowski's real contribution is upstream of the code. Knowing what to visualize and how to make interaction teach something rather than just look impressive is the rare part
This was for https://non.io, which is something I had purposefully written to be as fast as possible (hand wrote all the comopnents, didnt even use react).
I've been considering creating a skill / utility to do this based on learnings from the speedups - would others find this kind of thing useful?
It still feels pretty slow, and I see some low-hanging fruit:
- in safari, every image is loaded twice, .heic and .webp
- default.png is re-downloaded 31 times, uncached
- images below the fold are request immediately, lazy loading would avoid that [1]
But most important, you have 229 requests for tiny files being served over HTTP 1.1. Without GZIP. From a pretty slow server - 700ms+ to download the main json data. Bundling your JS, or enabling HTTP2 or QUIC/HTTP3 alone would massively improve performance.
If this is the result of days of autoresearch, it's not really anything to celebrate.
Why is it so slow? Like clicking around this is a very simple site, it seems like the fade in and fade out, besides being jarring and annoying, is just adding load time.
And there are plenty of things you did not opimize for. Like making sure certain dimensions are already known to the dom renderer so that the layout doesn't jump around.
Or progressive images so that it doesn't just popup suddenly.
An annoyance: clicking an image transitions the tiny image to a bigger size, then it gets replaced with a larger image file. The result is: the image slides bigger (but blurred), then immediately disappears, and reloads slowly from the top down. It's an annoying flash, and the slide to a bigger size was a waste of time.
That seriously needs some optimising. For example: on click, could the bigger image be inserted behind the small one so it's hidden; on load of the bigger image, hide the smaller image; then do the slide-bigger transition with both images together?
At this point, I'm half convinced that someone with $100k in api spend of Fable tokens can create a better replacement for X11/Wayland, and patch all common open source apps to leverage it.
Likely, but more the point I'm making is we're now approaching a point where this kind of issue is solvable with resources in reach of a small company / wealthy individual.
In their example, they list btn-* as the selector catch all for .btn-primary|secondary|danger. I can't help but think why not just do, .btn.primary for the class name, and just target .btn with the selector?
Don't get me wrong, I appreciate the convenience of this, but I do worry about selector slowdown with what will effectively turn into a regex at some point. I'm dubious that this is needed.
I have noticed it's pretty common for devs used to CSS-in-JS to imagine classes as being an exact 1:1 mapping to a specific DOM element, so maybe this is aiming to simplify things for that crowd? I guess similarly to BEM class names from a several years back.
Personally I prefer `class="btn primary"` over `class="btn-primary"` just because it aligns conceptually with what "class" literally means, but I have run into folks that would think it's confusing that there's no top-level .primary rule.
For performance... ehhhh yeah. Nesting and :has() already let you easily slow stuff down if you're not careful. Adding just wildcards, even if they're as restrictive as the blog post talks about, is still going to hang another easy-to-reach footgun on the proverbial wall.
> Personally I prefer `class="btn primary"` over `class="btn-primary"` just because it aligns conceptually with what "class" literally means, but I have run into folks that would think it's confusing that there's no top-level .primary rule.
Especially now that there is native nesting, these prefix selectors really seem like the wrong way to go.
The proposed mixin support in CSS will offer a way to reference both button.primary and a.primary without having to duplicate the styles in both, but it’s still in draft status.
Tailwind at this point is a coding language and a brand with no direct product. It's a great acquisition when you want mind share and developer love, and are OK with there being no revenue involved. Shopify is a solid match for that.
reply