LLM agents play Pokemon showdown
grabnero
25 points
14 comments
October 22, 2026
Related Discussions
Found 5 related stories in 113.2ms across 10,209 title embeddings via pgvector HNSW
- Multi-Agents LLM Financial Trading Framework fittingopposite · 115 pts · September 08, 2026 · 52% similar
- LLMs are real, AI is fake danaris · 69 pts · September 12, 2026 · 50% similar
- Show HN: Clawfight.ai MCP-driven agentic game play wesleyhales · 13 pts · September 11, 2026 · 49% similar
- LLMs as a Cognitive Virus canjobear · 239 pts · September 05, 2026 · 49% similar
- A1ex: A simple LLM coding agent in Lua uneven9434 · 24 pts · September 14, 2026 · 49% similar
Discussion Highlights (5 comments)
deadbabe
Jev will give better results on this kind of thing.
pgn
what was the selection for the teams? random per round? how does it play against people in ladder?
doctorpangloss
Pokémon Showdown isn't a game so much as it is competitive Wiki reading.
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.
jfreds
Did you try out different reasoning levels? Wondering if messaging is basically a proxy for additional reasoning