Evals for Every Language - Language vmw: leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 43.07 |
| 2 | Claude Opus 4.7 | 38.07 |
| 3 | GPT-5.5 | 37.15 |
| 4 | Gemini 2.5 Pro | 36.03 |
| 5 | Gemini 3.1 Flash Lite | 35.99 |
| 6 | Claude Sonnet 4.5 | 34.2 |
| 7 | GPT-5.2 | 34.06 |
| 8 | GPT-5.4 | 31.89 |
| 9 | DeepSeek V3.2 Exp | 31.76 |
| 10 | GPT-5.3 | 31.52 |
| 11 | Grok 4.20 | 31.37 |
| 12 | Llama 3.1 70B Instruct | 31.28 |
| 13 | Gemini 2.5 Flash | 30.3 |
| 14 | Command A | 30.28 |
| 15 | Claude Sonnet 4.6 | 30.17 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-vmw · How It Works · Data refreshed daily, snapshot 2026-09-05.