Benchmarks · reasoning
AI Chess Leaderboard (Dubesor)
LLMs play full chess games against each other from the move list, in reasoning and continuation modes combined; Elo within this LLM-only pool, models with 10+ games.
203 models · 203 results · updated 2026-10-31 · dubesor.de ↗
Leaderboard
Gemini 3.1 Pro high
More from this benchmark
3 published measuresScore vs any published measure3 measures: sub-scores, splits, counts
Breakdown
darker is better within each column · click a column to sort| Model | ||
|---|---|---|
| Gemini 3.1 Pro high | 84.0 | 82.0 |
| Gemini 3 Pro high | 100.0 | 75.0 |
| Gemini 3.8 Flash max | 105.0 | 39.0 |
| Gemini 3.7 Flash max | 111.0 | 38.0 |
| Gemini 3.6 Flash xhigh | 103.0 | 47.0 |
| Gemini 3.5 Flash xhigh | 84.0 | 63.0 |
| GPT-6 Astra max | 84.0 | 47.0 |
| Gemini 4 Argon max | 98.0 | 35.0 |
| GPT-5.1 Codex max | 98.0 | 37.0 |
| GPT-5 Codex high | 139.0 | 36.0 |
| Gemini 3 Flash high | 69.0 | 105.0 |
| Opus 5.5 max, fallback | 94.0 | 38.0 |
| GPT-5.5 xhigh | 88.0 | 53.0 |
| GPT-5 high | 93.0 | 54.0 |
| o3 high | 79.0 | 71.0 |
| GPT-6.1 Sol max | 92.0 | 39.0 |
| GPT-6.1 Terra max | 99.0 | 35.0 |
| Fable 5 max | 88.0 | 62.0 |
| GPT-5.6 Sol max | 70.0 | 68.0 |
| GPT-4.5 | 155.0 | 35.0 |
| DeepSeek V4 Flash 0731 max | 97.0 | 46.0 |
| Kimi K3 max | 103.0 | 37.0 |
| GPT-5.1 high | 101.0 | 39.0 |
| DeepSeek V4 Pro 0813 max | 119.0 | 26.0 |
| Qwen 3.8 Max | 101.0 | 36.0 |
All results
203 rows| # | Model | Effort | Score |
|---|---|---|---|
| 1 | Gemini 3.1 Pro | high | 1856 ±84 |
| 2 | Gemini 3 Pro | high | 1834 ±100 |
| 3 | Gemini 3.8 Flash | max | 1809 ±105 |
| 4 | Gemini 3.7 Flash | max | 1807 ±111 |
| 5 | Gemini 3.6 Flash | xhigh | 1750 ±103 |
| 6 | Gemini 3.5 Flash | xhigh | 1715 ±84 |
| 7 | GPT-6 Astra | max | 1640 ±84 |
| 8 | Gemini 4 Argon | max | 1630 ±98 |
| 9 | GPT-5.1 Codex | max | 1611 ±98 |
| 10 | GPT-5 Codex | high | 1602 ±139 |
| 11 | Gemini 3 Flash | high | 1583 ±69 |
| 12 | Opus 5.5 | max | 1570 ±94 |
| 13 | GPT-5.5 | xhigh | 1568 ±88 |
| 14 | GPT-5 | high | 1514 ±93 |
| 15 | o3 | high | 1496 ±79 |
| 16 | GPT-6.1 Sol | max | 1493 ±92 |
| 17 | GPT-6.1 Terra | max | 1490 ±99 |
| 18 | Fable 5 | max | 1465 ±88 |
| 19 | GPT-5.6 Sol | max | 1459 ±70 |
| 20 | GPT-4.5 | 1454 ±155 | |
| 21 | DeepSeek V4 Flash 0731 | max | 1444 ±97 |
| 22 | Kimi K3 | max | 1444 ±103 |
| 23 | GPT-5.1 | high | 1442 ±101 |
| 24 | DeepSeek V4 Pro 0813 | max | 1432 ±119 |
| 25 | Qwen 3.8 Max | max | 1430 ±101 |
| 26 | Grok 4 | high | 1415 ±118 |
| 27 | Gemini 3.1 Flash-Lite | high | 1403 ±78 |
| 28 | Grok 4.20 | xhigh | 1362 ±89 |
| 29 | DeepSeek V4.1 Flash 1017 | max | 1359 ±104 |
| 30 | DeepSeek V4.1 Flash | max | 1353 ±104 |