Alexandria - Human Dialectness Rating (Tunisian Arabic): leaderboard
Metric: Dialectness and fluency (1-5; native Tunisian speakers rate whether the English-to-dialect translation sounds like a native speaker of the target dialect). Source: arxiv.org. Saturation forecast: Around February 2027. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Preview) (Medium) | 3.24 |
| 2 | Gemini 3 Flash (Preview) (Minimal) | 3.2 |
| 3 | c4ai-command-a-03-2025 | 2.85 |
| 4 | GPT-OSS-120B | 2.58 |
| 5 | Gemma 3 27B (IT) | 2.3 |
| 6 | aya-expanse-32B | 2.22 |
Interactive version: theaggregate.ai/benchmark?slug=alexandria-human-dialectness-rating-tunisian-arabic · How It Works · Data refreshed daily, snapshot 2026-09-26.