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: years away from saturation. 220 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-6 (Max) | 72 |
| 2 | GPT-5.5 Pro (xHigh) | 64 |
| 3 | GPT-5.4 Pro (xHigh) | 58.6 |
| 4 | Gemini 3.1 Pro (Preview) | 55 |
| 5 | GPT-5.6 Sol (Max) | 55 |
| 6 | GPT-5.5 (xHigh) | 54 |
| 7 | GPT-5.6 Terra (Max) | 54 |
| 8 | Gemini 3.5 Flash (High) | 50 |
| 9 | Gemini 3.1 Pro (Preview) (High) | 49 |
| 10 | GPT-5.2 (xHigh) | 49 |
| 11 | Gemini 3.7 Flash (High) | 47 |
| 12 | Claude Fable 5.1 (Max) | 47 |
| 13 | DeepSeek V4 Pro (0813) (Max) | 47 |
| 14 | GPT-5.4 (xHigh) | 44 |
| 15 | Gemini 3.5 Flash (Minimal) | 43 |
Interactive version: theaggregate.ai/benchmark?slug=chess-puzzles-epoch-ai · How It Works · Data refreshed daily, snapshot 2026-09-05.