Evals for Every Language - Akan (ak): leaderboard
Average score across evaluation tasks in Akan (ak), from Evals for Every Language.
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Around January 2027. 66 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 68.58 | #58 |
| 2 | Claude Opus 4.8 | 65.51 | #37 |
| 3 | Claude Opus 4.7 | 64.39 | #49 |
| 4 | GPT-5.3 | 63.27 | #103 |
| 5 | Grok 4.20 | 62.27 | #175 |
| 6 | GPT-5.5 | 60.98 | #31 |
| 7 | Claude Sonnet 4.5 | 58.95 | #142 |
| 8 | Gemini 2.5 Pro | 57.89 | #152 |
| 9 | GPT-5.4 | 57.84 | #79 |
| 10 | Claude Sonnet 4.6 | 57.83 | #90 |
| 11 | GPT-5 | 56.9 | #98 |
| 12 | GPT-5.2 | 55.76 | #110 |
| 13 | GPT-5.1 | 55.57 | #135 |
| 14 | DeepSeek V3.2 Exp | 55.02 | #234 |
| 15 | Gemini 3.1 Flash Lite | 54.47 | #191 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-ak · How It Works · Data refreshed daily, snapshot 2026-10-20.