gaslight clock

Deepseek has a recognisable clock now and then, too. They both mix up the current time, though.

By far the best, but still off. These three were loaded in the same order as i post them:

Haha, I found myself thinking the same thing, and then caught myself, realizing all the other LLMs on this page had lowered the bar immensely for what I'm considering impressive.

It's the expected result if your big ol' artificial intelligence wannabe is ultimately just a stochastic word combinator.

if every single token is, at the end, chosen by random dice roll (and they are) then this is exactly what you'd expect.

This is hilariously sad how LLMs aren’t able to be consistent from one prompt to another.

Typically that's configurable. Like for a chatbot, you'd want it to give the same/similar results for a given question, where with a character creator, you might want the results to vary so you can re-run until you get something you like.

Of course that wouldn't be as funny here.

When I tried it Kimi K2 was surprisingly consistent and not even as bad as the others. Occasionally the numbers or hands (I couldn't really tell which) were possitioned a bit off, for example the seconds hand will appear to be horizontal but the 9 or 3 will be slightly below or slightly above the hand. But whoever can center a div may throw the first stone, and it's not going to be me for sure

You can click on the button in the top right corner (with a question mark) to have explanations. The clocks are refreshed every minute

qwen 2.5 is absolutely pants on head ridiculous compared to gpt5 when I'm looking at it right now.

So far, I’d give qwen the prize for most artistic impression of a clock.

Kimi K2 appears to consistently get it right.

I don't know if it's actually related, but I've read that asking people to draw a clock face is a simple way to identify some brain problems

Quick screening for dementia, according to this

Edit: I guess this means most of the AI has 'Conceptual Deficits', pretty accurate lol

I'm guessing it's an easy metric to compare benchmarks. "Write a clock".

Depends on when you load it. They refresh every minute. In the first one I got, Deepseek's was almost functional, but Haiku had one that was surprisingly good.

I've never seen any of the OpenAI models come up with anything that was more than completely broken.

I thought the same, until it refreshed and it got worse and it refreshed and it was different but still bad. These three were loaded in the order as posted:

Each model is allowed 2000 tokens to generate its clock. Here is its prompt: 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.

are you using the same prompt?