Model profile
Claude Opus 5 playing video games
3 GAMESGames played: Go (board game) · Heads-up Poker · NetHack & roguelikes (BALROG). Each game below is scored on its own terms.
Achievements
No achievement records yet.
Memorable failures
No failures on record. That means none have been sourced yet, not that none happened.
Leaderboard placements one row per game, never combined
| Game / board | Rank | Result | Status | As of |
|---|---|---|---|---|
| BALROG: BabyAI, Crafter, TextWorld, BabaIsAI, MiniHack, NetHack as Claude Opus 5 (max) | 2 | 63.4 Avg progress % | Verified | 2026-09-20 |
| Kaggle Game Arena: Go as Claude Opus 5 | 3 | 1,578 Elo | Verified | 2026-10-15 |
| Kaggle Game Arena: Heads-up Poker as Claude Opus 5 | 5 | 10 Mean BB/100 | Verified | 2026-10-08 |
Model versions seen
This Week in AI vs Games
A short weekly email with new records, memorable failures and runs to watch, every claim linked to its source.
Newsletter launching soon. Follow passports or subscribe by feed for now: site-wide Atom feed · passports to follow.
Latest issue: Issue #1: The Ender Dragon is down