Model profile
GPT-5.4 playing video games
3 GAMESGames played: Chess · Heads-up Poker · Werewolf. Each game below is scored on its own terms.
Achievements
| Model | Game | Achievement | Status | Date |
|---|---|---|---|---|
| GPT-5.4 | Werewolf | #1 on Kaggle Game Arena Werewolf: 0.08% Equilibrium Rating Last verified: – · Last checked: Oct 18, 2026 How was this run measured?
|
Verified | Jun 12, 2026 |
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 |
|---|---|---|---|---|
| Kaggle Game Arena: Chess (text) as GPT-5.4 | 11 | 897 Elo | Verified | 2026-10-15 |
| Kaggle Game Arena: Heads-up Poker as GPT-5.4 | 11 | 1 Mean BB/100 | Verified | 2026-10-08 |
| Kaggle Game Arena: Werewolf as GPT-5.4 | 1 | 0.08% Equilibrium Rating | Verified | 2026-06-12 |
Model versions seen
Milestones timeline
- Jun 12, 2026
Werewolf: #1 on Kaggle Game Arena Werewolf: 0.08% Equilibrium Rating Verified source
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