Evals for Every Language - Language ms: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Flash Lite | 79.05 |
| 2 | Gemini 3.1 Pro (Preview) | 78.68 |
| 3 | Claude Sonnet 4.5 | 77.74 |
| 4 | Gemini 2.5 Pro | 77.08 |
| 5 | Claude Sonnet 4 | 76.92 |
| 6 | GPT-5 | 75.42 |
| 7 | Claude Sonnet 4.6 | 75.27 |
| 8 | GPT-5.3 | 75.06 |
| 9 | GPT-5.4 | 74.3 |
| 10 | Grok 4.20 | 74.04 |
| 11 | GPT-5.2 | 73.94 |
| 12 | Claude Opus 4.7 | 73.55 |
| 13 | GPT-5.1 | 73.35 |
| 14 | GPT-5.5 | 73.27 |
| 15 | Qwen 3.6 Plus | 73 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-ms · How It Works · Data refreshed daily, snapshot 2026-09-05.