ChessBench - Elo: leaderboard
Monte Carlo Elo rating from ChessBench games, anchored to fixed reference players spanning the strength range.
Metric: Elo. Source: chessbench.ai. Saturation forecast: Around November 2027. 72 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 1755 | #6 |
| 2 | GPT-6.1 Sol | 1562 | #7 |
| 3 | Gemini 3.8 Flash | 1527 | #30 |
| 4 | Gemini 4 Argon | 1485 | #9 |
| 5 | GPT-5.5 | 1349 | #33 |
| 6 | Claude Opus 5.5 | 1349 | #3 |
| 7 | GPT-6 Sol | 1303 | #19 |
| 8 | Gemini 3.7 Flash | 1265 | #24 |
| 9 | GPT-6.1 Terra | 1251 | #15 |
| 10 | GPT-5.6 Sol | 1218 | #22 |
| 11 | GPT-5.1 | 1202 | #145 |
| 12 | Muse Spark 1.3 | 1201 | #31 |
| 13 | Gemini 3.1 Pro (Preview) | 1192 | #64 |
| 14 | Gemini 3.6 Flash | 1192 | #57 |
| 15 | Muse Spark 1.1 | 1146 | #55 |
Interactive version: theaggregate.ai/benchmark?slug=chessbench-elo · How It Works · Data refreshed daily, snapshot 2026-11-05.