ChessBench — Mate in 2
Games · 2026-03-22
ChessBench Mate in 2 contains 30 fixed Lichess positions in which the model must return the exact expected three-ply UCI mating line. The track is stratified evenly across 800–1200, 1200–1600, and 1600–2000 puzzle-rating buckets.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 3.8 Flash | 86.7 |
| Claude Opus 5.5 | 76.7 |
| GPT-6.1 Sol | 76.7 |
| GPT-6 Sol | 76.7 |
| Gemini 4 Argon | 73.3 |
| Claude Haiku 5.5 | 56.7 |
| Gemini 3 Flash Preview | 53.3 |
| GPT-6 Luna | 53.3 |
| Grok 4.6 | 53.3 |
| Claude Fable 5.1 | 40.0 |
| Claude Sonnet 5.5 | 30.0 |
| Haiku 4.5 | 0.0 |