Sonnet 4.5 does it flawless. Tried 8 times.
The catch was that it was limited to 2000 tokens, i.e. the results get cut off once it hits that.
AI World Clocks
391–400 of 404 posts
Re: AI World Clocks
#392I’m very curious about the monthly bill for such a creative project, surely some of these are pre rendered?
Napkin math: 9 AIs × 43,200 minutes = 388,800 requests/month 388,800 requests × 200 tokens = 77,760,000 tokens/month ≈ 78M tokens Cost varies from 10 cents to $1 per 1M tokens. Using the mid-price, the cost is around $50/month. --- Hopefully, the OP has this endpoint protected - https://clocks.brianmoore.com/api/clocks?time=11:19AM
It was limited to 2,000 tokens each. I assume it usually hit that. So could be closer to 777M. assuming they didn't just cache it and just start rotating after a day or two..
Re: AI World Clocks
#393Earlier quoted context omitted.
i read that the OP limited the output to 2000 tokens.
^ this! there's a lot of clocks to generate so I've challenged it to stick to a small(er) amount of code
I wonder if you would get better results if you tell the LLM there's a token limit in the prompt.
something like "You only have 1000 tokens. Generate an analog clock showing ${time}, with a CSS animated second hand. Make it responsive and use a white background. Return ONLY the HTML/CSS code with no markdown formatting"
Re: AI World Clocks
#394Think this is just prompt eng tbh. One shot Haiku 3.5 ( https://claude.ai/share/66c17968-485e-4d15-974b-4f6958e1e2fd ) decent looking too. Got it to work on gpt 3.5T w modified prompt (albeit not as good - https://pastebin.com/gjEVSEcJ ) `single html file, working analog clock showing current time, numbers positioned (aligned) correctly via trig calc (dynamic), all three hands, second hand ticks, 400px, clean AF aest…
The catch was that it was limited to 2000 tokens, i.e. the results get cut off once it hits that.
As in, no more than 2K output tokens?
Responses above are about ~1K.
Re: AI World Clocks
#395would be gr8t to also see the prompt this was done with
The ? has "Create HTML/CSS of an analog clock showing ${time}. Include numbers (or numerals) if you wish, and have a CSS animated second hand. Make it responsive and use a white background. Return ONLY the HTML/CSS code with no markdown formatting."
Hmm nothing fancy then, but perhaps with tubing results will vary.
I hate prompt discovery (not engineering this thing!), but it actually matters.
Re: AI World Clocks
#396Re: AI World Clocks
#397Earlier quoted context omitted.
Maybe LLMs always fail to generalize outside their data set, and it’s just less noticeable with written language.
Most image models are diffusion models, not LLMs, and have a bunch of other idiosyncrasies. So I suspect it's more that lessons from diffusion image models don't carry over to text LLMs. And the Image models which are based on multi-mode LLMs (like Nano Banana) seem to do a lot better at novel concepts.
But the clocks in this demo aren't images.
Re: AI World Clocks
#398Earlier quoted context omitted.
Most image models are diffusion models, not LLMs, and have a bunch of other idiosyncrasies. So I suspect it's more that lessons from diffusion image models don't carry over to text LLMs. And the Image models which are based on multi-mode LLMs (like Nano Banana) seem to do a lot better at novel concepts.
But the clocks in this demo aren't images.
Yes, but they are reasoning within their dataset, which will contain multiple example of html+css clocks.
They are just struggling to produce good results because they are language models and don’t have great spatial reasoning skills, because they are language models.
Their output normally has all the elements, just not in the right place/shape/orientation.
Re: AI World Clocks
#399Earlier quoted context omitted.
Without AI assistance, this should take ~10–15 minutes for a human. Maybe add 5 minutes if you're not allowed to use d3.
I would not even know how to draw a circle with CSS to be honest.
A div with width = height = border-radius (or width = height with border-radius:50%)
Re: AI World Clocks
#400Earlier quoted context omitted.
Clock drawing is widely used as a test for assessing dementia. Sometimes the LLMs fail in ways that are fairly predictable if you're familiar with CSS and typical shortcomings of LLMs, but sometimes they fail in ways that are less obvious from a technical perspective but are exactly the same failure modes as cognitively-impaired humans. I think you might have stumbled upon something surprisingly profound. https://www…
Figure 6 with the square clock would be a cool modern art piece.
I have had this thought of a slow-moving mechanical simulation of a chaotic triple pendulum as a clock hand for a very long time..
Or maybe something like https://www.youtube.com/watch?v=dhZxdV2naw8