Model profile
GPT-4o playing video games
5 GAMESGames played: Factorio · Multi-game (Orak: 12 games) · Pokémon Crystal · VideoGameBench Lite suite (3 games) · VideoGameBench suite (10 test games). Each game below is scored on its own terms.
Achievements
No achievement records yet.
Memorable failures
| Model | Game | Achievement | Status | Date |
|---|---|---|---|---|
| GPT-4o (2024-08-06) | Pokémon Crystal | Memorable fail 0.9% of Pokémon Crystal completed on VideoGameBench (test set, real-time) Picked Cyndaquil as its starter, then lost track of what it was supposed to do. Last verified: Oct 18, 2026 · Last checked: Oct 18, 2026 How was this run measured?
|
Verified | May 23, 2025 |
Leaderboard placements one row per game, never combined
| Game / board | Rank | Result | Status | As of |
|---|---|---|---|---|
| Factorio Learning Environment as GPT-4o | 3 | 87,599 Production score | Verified | 2026-05-28 |
| Orak (KRAFTON AI): 12 video games as GPT-4o (2024-08-06) | 2 | 4.3 Avg rank (lower is better) | Verified | 2025-06-09 |
| VideoGameBench Lite (game pauses while the model thinks) as GPT-4o (2024-08-06) | 1 | 1.6% Overall score | Verified | 2025-05-23 |
| VideoGameBench (real-time) as GPT-4o (2024-08-06) | 2 | 0.09% Overall score | Verified | 2025-05-23 |
Model versions seen
Milestones timeline
- May 23, 2025
✗ Pokémon Crystal: 0.9% of Pokémon Crystal completed on VideoGameBench (test set, real-time) Verified source
This Week in AI vs Games
One short email every Sunday: new records, memorable failures and runs to watch. Free. Every claim is linked to its source.
Latest issue: Issue #1: The Ender Dragon is down