LLM teammates that talk freely win 64% of Pokémon doubles battles
LLM Agents Play Pokemon Showdown
A researcher wrapped Pokémon Showdown's 2v2 format into a Gym-style environment to test whether LLM agents can coordinate. Two Qwen3.8-27B teammates that could chat freely beat a max-damage bot 64% of the time, versus 42% with no talk — though free talk burned 978k prompt tokens per battle, 3.8x the silent baseline. The setup also cleanly separates model strengths, with the 27B team winning 96% against 2B opponents.
Talking helps. Teammates who could talk freely won 64% of their battles against the max-damage bot, while teammates who couldn't talk at all won 42%.
- minimaxir
As a tangent, there has recently been drama with AI and Pokemon Showdown because someone trained a bot (using standard reinforcement learning + a rewrite of the Pokemon battle engine to Rust for max self-play parallelism) that is apparently good enough to get to the top of the leaderboard.
- cryzinger
> Pokémon doubles is a game-theory environment without full information. The opponents' sets are hidden[...]
I'd be curious to see how a model fares in VGC rules with open team sheets, since that's where you swap out information-hiding games for good old-fashioned mind games :P
> [...] and damage is random [...]
Not to nitpick, but since damage isn't truly random, did it take possible damage rolls into effect? Being able to tell whether your opponent has a damage-boosting item/ability matters a lot!
- jfreds
Did you try out different reasoning levels? Wondering if messaging is basically a proxy for additional reasoning