
The metric for any given puzzle is a binary "Did it solve it?", with no consideration for how elegant the process is/isn't, or how close it got.
Results will use:
My already submitted version 3 solution, currently been testing by Mira on the January puzzles.
Mira's solution, currently being developed.
I don't know exactly how Mira's solution will be tested (how many puzzles), so:
Resolves YES if Mira's prompt is better than mine (higher solve percentage, + common sense of intent, to avoid a successful solve on only a single attempted puzzle counting as 1/1 = 100% etc).
Resolves NO if my prompt is better than Mira's (higher solve percentage, + common sense of intent), or if Mira does not finish a solution or test it on any puzzles. Also resolves NO if the prompts objectively tie.
Resolves N/A if neither Yes or No makes sense.
May resolve early if all testing is finished early. May extend if testing needs longer, for reasons agreed to be sensible.
🏅 Top traders
# | Name | Total profit |
---|---|---|
1 | Ṁ6 | |
2 | Ṁ6 | |
3 | Ṁ6 | |
4 | Ṁ5 | |
5 | Ṁ5 |