Chess Puzzles (Epoch AI) — leaderboard
Chess puzzle-solving benchmark evaluated by Epoch AI. Tests tactical pattern recognition, calculation depth, and strategic reasoning across a range of puzzle difficulties.
Metric: Accuracy (%). Source: epoch.ai. Status: saturation imminent. 57 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 Pro (xHigh) | 64 |
| 2 | GPT-5.4 Pro (xHigh) | 58.6 |
| 3 | Gemini 3.1 Pro (Preview) | 55 |
| 4 | GPT-5.6 Sol (Max) | 55 |
| 5 | GPT-5.5 (xHigh) | 54 |
| 6 | GPT-5.6 Terra (Max) | 54 |
| 7 | Gemini 3.5 Flash (High) | 50 |
| 8 | GPT-5.2 (xHigh) | 49 |
| 9 | GPT-5.4 (xHigh) | 44 |
| 10 | Claude Fable 5 (Max) | 41 |
| 11 | GPT-5.2 (Medium) | 40 |
| 12 | GPT-5.2 (High) | 40 |
| 13 | GPT-5.6 Luna (Max) | 40 |
| 14 | Kimi K3 (Max) | 39 |
| 15 | Gemini 3 Flash (Preview) | 38 |
Interactive version: theaggregate.ai/benchmark?slug=chess-puzzles-epoch-ai · How the rankings work · Data refreshed daily, snapshot 2026-07-22.