I've been trying to get some language models to paint one stroke at a time for a few months now. I thought this community would be interested to see the results.
The article runs through my findings, and there's a linked technical rundown of how the app was built. There's also an interactive gallery [0] of my attempts. You can point an agent at the API docs [1], and they might (ymmv) do a painting themselves.
I would say how, but I am not your friend and here in the 2030s, no one can afford to give anything valuable for free to a stranger. Be glad of the advice, of which you'd be wise to make much more than you will.
You may enjoy
* "The last six months in LLMs, illustrated by pelicans on bicycles" https://simonwillison.net/2025/Jun/6/six-months-in-llms/ (https://news.ycombinator.com/item?id=44215352 | 962 points | 11 months ago | 239 comments)
* "Using “underdrawings” for accurate text and numbers" https://samcollins.blog/underdrawings/ (https://news.ycombinator.com/item?id=47977990 | 379 points | 9 days ago | 138 comments)
I've been trying to get some language models to paint one stroke at a time for a few months now. I thought this community would be interested to see the results.
The article runs through my findings, and there's a linked technical rundown of how the app was built. There's also an interactive gallery [0] of my attempts. You can point an agent at the API docs [1], and they might (ymmv) do a painting themselves.
[0] https://www.liamlaverty.com/paint-by-language-model/ [1] https://www.liamlaverty.com/paint-by-language-model/draw/api
You are desperately misguided.
I would say how, but I am not your friend and here in the 2030s, no one can afford to give anything valuable for free to a stranger. Be glad of the advice, of which you'd be wise to make much more than you will.
Good attempt. Compared to diffusion, these paintings look more like they were created by humans.
LLMs can draw (play music, write books), but they imitate, not create.