[dead]
AI World Clocks
291–300 of 404 posts
Re: AI World Clocks
#292hi, I made this. thank you for posting. I love clocks and I love finding the edges of what any given technology is capable of. I've watched this for many hours and Kimi frequently gets the most accurate clock but also the least variation and is most boring. Qwen is often times the most insane and makes me laugh. Which one is "better?"
If you're keeping all the generated clocks in a database, I'd love to see a Facemash style spin-off website where users pick the best clock between two options, with a leaderboard. I want to know what the best clock Qwen ever made was!
[dead]
Re: AI World Clocks
#293Earlier quoted context omitted.
How much engineering do prompt engineers do? Is it engineering when you add "photorealistic. correct number of fingers and teeth. High quality." to the end of a prompt? we should call them "prompt witch doctors" or maybe "prompt alchemists".
I write quite a lot of prompts, and the closest analogy that I can think of is a shaman trying to appease the spirits.
[dead]
Re: AI World Clocks
#294The more I look at it, the more I realise the reason for cognitive overload I feel when using LLMs for coding. Same prompt to same model for a pretty straight forward task produces such wildly different outputs. Now, imagine how wildly different the code outputs when trying to generate two different logical functions. The casings are different, commenting is different, no semantic continuity. Now maybe if I give detailed prompts and ask it to follow, it might follow, but from my experience prompt adherence is not so great as well. I am at the stage where I just use LLMs as auto correct, rather than using it for any generation.
Re: AI World Clocks
#295Re: AI World Clocks
#296hi, I made this. thank you for posting. I love clocks and I love finding the edges of what any given technology is capable of. I've watched this for many hours and Kimi frequently gets the most accurate clock but also the least variation and is most boring. Qwen is often times the most insane and makes me laugh. Which one is "better?"
Could you please change and adjust the positions of the titles (like GPT 5)? On Firefox Focus on iOS, the spacing is inconsistent (seems like it moves due to the space taken by the clock). After one or two of them, I had to scroll all the way down to the bottom and come back up to understand which title is linked to which clock.
Re: AI World Clocks
#297hi, I made this. thank you for posting. I love clocks and I love finding the edges of what any given technology is capable of. I've watched this for many hours and Kimi frequently gets the most accurate clock but also the least variation and is most boring. Qwen is often times the most insane and makes me laugh. Which one is "better?"
This is an awesome benchmark. Officially one of my favorites now. Thank you for making this.
Re: AI World Clocks
#298hi, I made this. thank you for posting. I love clocks and I love finding the edges of what any given technology is capable of. I've watched this for many hours and Kimi frequently gets the most accurate clock but also the least variation and is most boring. Qwen is often times the most insane and makes me laugh. Which one is "better?"
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…
Maybe explainable via the fact that these tests are part of the LLM training set?
Re: AI World Clocks
#299You should render it, show an image to the model and allow it to iterate. No person has to one-shot code without seeing what it looks like.
Re: AI World Clocks
#300hi, I made this. thank you for posting. I love clocks and I love finding the edges of what any given technology is capable of. I've watched this for many hours and Kimi frequently gets the most accurate clock but also the least variation and is most boring. Qwen is often times the most insane and makes me laugh. Which one is "better?"
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…
Conceptual deficit is a great failure mode description. The inability to retrieve "meaning" about the clock -- having some understanding about its shape and function but not its intent to convey time to us -- is familiar with a lot of bad LLM output.