Chess Bench LLM — leaderboard
Metric: Lichess Rating. Source: chessbenchllm.onrender.com. 102 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Human Expert | 3000 |
| 2 | Gemini 3.1 Pro (Preview) (High) | 2066 |
| 3 | Gemini 3.5 Flash (Medium) | 1957 |
| 4 | Gemini 3 Pro (Preview) (High) | 1803 |
| 5 | Gemini 3 Flash (Preview) (Medium) | 1647 |
| 6 | Gemini 3 Flash (Preview) (High) | 1635 |
| 7 | GPT-5.6 Sol (xHigh) | 1612 |
| 8 | GPT-5.6 Sol (High) | 1538 |
| 9 | Grok 4.1 Fast | 1405 |
| 10 | Claude Opus 4.6 (Medium) | 1403 |
| 11 | GPT-5.6 Sol (Medium) | 1392 |
| 12 | GPT-5.6 Luna (xHigh) | 1371 |
| 13 | O3 | 1336 |
| 14 | GPT-5.4 (High) | 1319 |
| 15 | GPT-5.1 (High) | 1313 |
Interactive version: theaggregate.ai/benchmark?slug=chess-bench-llm · How the rankings work · Data refreshed daily, snapshot 2026-07-22.