Evals for Every Language - Language war: leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 61.02 |
| 2 | Gemini 2.5 Pro | 60.79 |
| 3 | Claude Opus 4.7 | 58.17 |
| 4 | Gemini 2.5 Flash | 57.49 |
| 5 | Claude Opus 4.8 | 57.14 |
| 6 | Gemini 3.1 Flash Lite | 56.89 |
| 7 | GPT-4o | 56.85 |
| 8 | GPT-5.3 | 56.76 |
| 9 | Claude Sonnet 4.5 | 56.21 |
| 10 | GPT-5.1 | 55.25 |
| 11 | GPT-5.5 | 55.01 |
| 12 | Claude Sonnet 4 | 54.83 |
| 13 | Claude Sonnet 4.6 | 54.49 |
| 14 | GPT-5 | 53.54 |
| 15 | GPT-5.4 | 53.33 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-war · How It Works · Data refreshed daily, snapshot 2026-09-05.