RankJudge — leaderboard
Metric: Elo (All) (self-reported). Source: benchmarklist.com. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 1959 |
| 2 | GPT-5.5 | 1867 |
| 3 | Gemini 3 Flash (Preview) | 1724 |
| 4 | Claude Sonnet 4.6 | 1706 |
| 5 | Gemma 4 31B | 1695 |
| 6 | Qwen 3.6 Plus | 1671 |
| 7 | GLM-5.1 | 1637 |
| 8 | GPT-5.4 | 1631 |
| 9 | Qwen 3.5 397B A17B | 1571 |
| 10 | Claude Opus 4.7 | 1510 |
| 11 | Claude Haiku 4.5 | 1505 |
| 12 | Gemma 4 26B A4B | 1444 |
| 13 | Qwen 3.5 122B A10B | 1422 |
| 14 | Qwen 3.5 35B A3B | 1373 |
| 15 | GPT-5.4 Mini | 1342 |
Interactive version: theaggregate.ai/benchmark?slug=rankjudge · How the rankings work · Data refreshed daily, snapshot 2026-07-22.