The Metanym Game: leaderboard
The metanym game is a competitive word game for LLMs that measures structural intelligence against established cognitive-science constructs. No content is given in advance; the contestants create all of it -- a new kind of analogy test, analogical production falsifiable sentence by sentence, with no fixed test set to leak into training (contamination-resistant by construction). In the council-of-p
Metric: T (self-reported). Source: benchmarklist.com. Saturation forecast: Around April 2027. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.5 | 7 | #83 |
| 2 | Gemini 3.1 Pro (Preview) | 6.69 | #58 |
| 3 | Claude Opus 4 | 6.21 | #163 |
| 4 | Claude Opus 4.1 | 6.05 | #140 |
| 5 | Gemini 2.5 Flash | 5.77 | #245 |
| 6 | Claude Sonnet 4 | 5.3 | #199 |
| 7 | GPT-4.1 Mini | 4.76 | #355 |
| 8 | GPT-4.1 | 4.44 | #248 |
| 9 | GPT-4o (2024-08-06) | 3.49 | #336 |
| 10 | GPT-4.1 Nano | 3.22 | #731 |
| 11 | GPT-4o | 2.93 | #343 |
| 12 | GPT-4o Mini | 2.24 | #599 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=the-metanym-game · How It Works · Data refreshed daily, snapshot 2026-10-20.