Evals for Every Language - Sango (sg): leaderboard
Average score across evaluation tasks in Sango (sg), from Evals for Every Language.
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Around June 2027. 64 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 | 68.86 | #49 |
| 2 | Gemini 3.1 Pro (Preview) | 68.83 | #58 |
| 3 | GPT-5.5 | 65.89 | #31 |
| 4 | Qwen 3.6 Plus | 64.57 | #120 |
| 5 | Gemini 3.1 Flash Lite | 62.65 | #191 |
| 6 | Claude Opus 4.8 | 62.37 | #37 |
| 7 | Grok 4.20 | 59.15 | #175 |
| 8 | Gemini 2.5 Pro | 58.13 | #152 |
| 9 | Claude Sonnet 4.6 | 54.42 | #90 |
| 10 | GPT-5.4 | 51.74 | #79 |
| 11 | DeepSeek V3.1 | 48.37 | #264 |
| 12 | jamba-large-1.7 | 45.83 | #525 |
| 13 | Qwen 3.6 Flash | 44.72 | #214 |
| 14 | GPT-5 | 43.53 | #98 |
| 15 | GPT-5.2 | 42.93 | #110 |
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-sg · How It Works · Data refreshed daily, snapshot 2026-10-20.