Kaggle Corp Speak Arena: leaderboard
Kaggle human-vote pairwise arena rating models on translating 100 mundane sentences across 10 categories into corporate speak (passing responses are 5-100 words of plain text). Reports Bradley-Terry style arena ratings per model and reasoning effort.
Metric: Arena rating (human votes). Source: www.kaggle.com. Saturation forecast: Rough model projection: around 2027. 24 models tracked. Updated 20 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Claude Opus 5 | Max | 1704.91 | #26 |
| 2 | GLM-5.3 | Max | 1695.46 | #73 |
| 3 | Claude Opus 4.7 | Max | 1643.59 | #85 |
| 4 | Kimi K3 | Max | 1620.32 | #61 |
| 5 | Claude Opus 5.5 | Max | 1598.44 | #9 |
| 6 | Gemini 3.1 Pro (Preview) | High | 1571.16 | #101 |
| 7 | Gemini 3.8 Flash | High | 1567.23 | #43 |
| 8 | GPT-6.1 Sol | Max | 1556.7 | #15 |
| 9 | Claude Fable 5.1 | Max | 1552.52 | #20 |
| 10 | Grok 4.7 | xHigh | 1550.53 | #51 |
| 11 | Claude Sonnet 5.5 | Max | 1535.49 | #24 |
| 12 | Gemini 4 Argon | High | 1519.1 | #18 |
| 13 | GPT-6 Astra | Max | 1514.57 | #10 |
| 14 | Claude Sonnet 5 | High | 1486.29 | #118 |
| 15 | GPT-5.5 | High | 1485.39 | #54 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=kaggle-corp-speak-arena · How It Works · Data refreshed daily, snapshot 2026-11-20.