Evals for Every Language - Ganda (lg): leaderboard
Average score across evaluation tasks in Ganda (lg), 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) | 71.92 | #58 |
| 2 | Claude Opus 4.8 | 68.6 | #37 |
| 3 | GPT-5.5 | 67.21 | #31 |
| 4 | GPT-5.3 | 65.86 | #103 |
| 5 | GPT-5.1 | 64.99 | #135 |
| 6 | GPT-5.4 | 64.86 | #79 |
| 7 | GPT-5 | 64.22 | #98 |
| 8 | Claude Opus 4.7 | 63.37 | #49 |
| 9 | DeepSeek V3.1 | 62.69 | #264 |
| 10 | Grok 4.20 | 62.22 | #175 |
| 11 | Gemini 3.1 Flash Lite | 61.58 | #191 |
| 12 | Gemini 2.5 Pro | 59.29 | #152 |
| 13 | GPT-5.2 | 58.98 | #110 |
| 14 | Claude Sonnet 4.6 | 58.32 | #90 |
| 15 | Qwen 3.6 Plus | 56.53 | #120 |
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-lg · How It Works · Data refreshed daily, snapshot 2026-10-20.