cshdllr

I honestly don't know! My guess is that Sol is better calibrated for the hard questions that require more reasoning...but even still I'd expect it to be able to handle a simple game like this better than its sibling models.

Sadly I think most human Slime Soccer players have retired. But more seriously, the way this is hooked up is not realtime, so the game would require pauses while the model decided what to do. Maybe next I can try more of a turn based RPG which would more easily incorporate human players.

Thanks!

About three times a second each model gets a fixed system prompt plus the current game state as JSON, and replies with one action word (move_right, move_left, jump, idle). This is all run as a simulation since they couldn't reply fast enough for it to be realtime. Then the video is made after the fact from the logs.

That's fair. From what I've seen, people play multiple rounds. So the first round they don't quite know what a "top" score looks like and then they play again after seeing the distribution. Since the game is short enough, it kind of feels like that might be enough feedback.