Evals for Every Language - Language bho: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 68.98 |
| 2 | Claude Sonnet 4.5 | 68.42 |
| 3 | Claude Sonnet 4 | 68.31 |
| 4 | GPT-5.5 | 67.61 |
| 5 | GPT-5.2 | 67.51 |
| 6 | Gemini 3.1 Pro (Preview) | 67.37 |
| 7 | Claude Opus 4.8 | 67.32 |
| 8 | Claude Opus 4.7 | 67.27 |
| 9 | GPT-5.1 | 67.24 |
| 10 | GPT-5.3 | 67.1 |
| 11 | GPT-5.4 | 65.92 |
| 12 | Qwen 3.6 Plus | 65.66 |
| 13 | Claude Sonnet 4.6 | 65.55 |
| 14 | Gemini 3.1 Flash Lite | 65.5 |
| 15 | GPT-5 Mini | 65.24 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-bho · How It Works · Data refreshed daily, snapshot 2026-09-05.