RankJudge: leaderboard
As interactive LLM-based applications are created and refined, model developers need to evaluate the quality of generated text along many possible axes. For simpler systems, human evaluation may be practical, but in complicated systems like conversational chatbots, the amount of generated text can overwhelm human annotation resources. Model developers have begun to rely heavily on auto-evaluation,
Metric: Elo (All) (self-reported). Source: benchmarklist.com. Saturation forecast: Around June 2027. 21 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 1959 | #58 |
| 2 | Gemini 3 Flash | 1724 | #99 |
| 3 | Kimi K2.6 | 1713 | #104 |
| 4 | Claude Sonnet 4.6 | 1706 | #90 |
| 5 | Gemma 4 31B | 1695 | #200 |
| 6 | Qwen 3.6 Plus | 1671 | #120 |
| 7 | GLM-5.1 | 1637 | #97 |
| 8 | GPT-5.4 | 1631 | #79 |
| 9 | Qwen 3.5 397B A17B | 1571 | #146 |
| 10 | Claude Opus 4.7 | 1510 | #49 |
| 11 | Claude Haiku 4.5 | 1505 | #280 |
| 12 | Gemma 4 26B A4B | 1444 | #358 |
| 13 | Qwen 3.5 122B A10B | 1422 | #173 |
| 14 | Qwen 3.5 35B A3B | 1373 | #256 |
| 15 | GPT-5.4 Mini | 1342 | #212 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=rankjudge · How It Works · Data refreshed daily, snapshot 2026-10-20.