It made me smile, I'm glad they built it.
Random idea: now animate it.
Run the initial generate, keep that result, provide it back to another llm call: “user requested an elephant. you drew this. (object here) generate the next frame of an animation of this.”
Iterate the prompt 2-3 times for cool animated ascii art :)
_.-=-._
.-` '---` '-.
/ .-"-. \
| / _ \ |
\ \ (_) / /
'-._ '-=-'_.-'
'---'
_
_|=|__________
/ \
/ \
/__________________\
| ________ ___ |
| | | | ||
| | | | ||
| | | | ||
| | | | ||
| | | | ||
| | | | ||
| |______| |___||
|_________________|
( ) ( )
( ) ( )
( ) ( )
It’s quite clever.
Matching styles using vision models is not that easy, especially if you want to capture the core subject of the prompt.
ASCII Art for "bambi":
, ,
/////|
///// |
///// |
|~~~| | |
|===| |/|
| B |/| |
| A / | |
| M | |
| B | |
| I | /
|===|/
'---'
/ \
/ \
/ .-""-. \
( / \ )
\/ \/
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
/ \
/ \ ___
_/ \_
/ \
| |
| |
| |
\_______/
/ \
/ \
/ \
/ \
/ \
| ___ |
| (___) |
| |
\ /
\_______/