Judgemark v2.1: leaderboard
Evaluates LLM-as-judge quality: stability (self-consistency), separability (ability to distinguish model quality), and correlation with human preferences.
Metric: Judgemark Score (0-100). Source: eqbench.com. Saturation forecast: Estimated already saturated. 47 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 94.08 | #90 |
| 2 | Claude Opus 4.6 | 92.93 | #63 |
| 3 | Gemini 3 Pro (Preview) | 86.58 | #66 |
| 4 | GLM-5 | 85.58 | #141 |
| 5 | Qwen 3.5 397B A17B | 85.45 | #146 |
| 6 | Gemini 3 Flash (Preview) | 85.38 | #82 |
| 7 | GPT-5.4 | 84.98 | #79 |
| 8 | Claude Sonnet 4.5 | 84.56 | #142 |
| 9 | Qwen 3.5 27B | 84.31 | #187 |
| 10 | GPT-5.2 | 82.01 | #110 |
| 11 | Claude Sonnet 4 | 81.99 | #199 |
| 12 | Qwen 3.5 Flash (02-23) | 81.85 | #185 |
| 13 | Qwen 3.5 35B A3B | 81.39 | #256 |
| 14 | GPT-5.4 Mini | 81.14 | #212 |
| 15 | Qwen 3.5 122B A10B | 81.03 | #173 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=judgemark-v2-1 · How It Works · Data refreshed daily, snapshot 2026-10-20.