Featherbench - Rubric Quality: leaderboard
Metric: Rubric Score (0-10). Source: reinvently.co.uk. Saturation forecast: Estimated already saturated. 25 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Kimi K3 | 9.5 |
| 2 | Claude Fable 5.1 | 9.5 |
| 3 | Claude Opus 5.5 | 9.5 |
| 4 | Claude Opus 5 | 9.4 |
| 5 | Claude Fable 5 | 9.3 |
| 6 | GLM-5.3 | 9.3 |
| 7 | GLM-5.3 Flash | 9.2 |
| 8 | Claude Sonnet 4.6 | 8.9 |
| 9 | Gemini 3.8 Flash | 8.9 |
| 10 | GPT-5.6 Sol | 8.8 |
| 11 | Claude Sonnet 5 | 8.8 |
| 12 | Gemini 3.6 Flash | 8.8 |
| 13 | Gemini 3.7 Flash | 8.8 |
| 14 | Grok 4.6 | 8.8 |
| 15 | DeepSeek V4.1 Flash | 8.8 |
Interactive version: theaggregate.ai/benchmark?slug=featherbench-rubric-quality · How It Works · Data refreshed daily, snapshot 2026-09-23.