Evals for Every Language - Translation To: leaderboard
LanguageBench Translation To spBLEU averaged across available languages and scaled to 0-100. Translation uses spBLEU only, without mixing chrF; higher is better.
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Around 2029. 71 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 30.27 | #58 |
| 2 | Gemini 2.5 Pro | 29.29 | #152 |
| 3 | Claude Opus 4.7 | 27.85 | #49 |
| 4 | Gemini 2.5 Flash | 27.57 | #245 |
| 5 | GPT-5.5 | 27.57 | #31 |
| 6 | Claude Opus 4.8 | 26.77 | #37 |
| 7 | Claude Sonnet 4.5 | 26.67 | #142 |
| 8 | Gemini 3.1 Flash Lite | 26.61 | #191 |
| 9 | GPT-5.4 | 25.76 | #79 |
| 10 | Qwen 3.6 Plus | 25.21 | #120 |
| 11 | Claude Sonnet 4 | 24.85 | #199 |
| 12 | GPT-5.3 | 24.48 | #103 |
| 13 | GPT-5.2 | 24.07 | #110 |
| 14 | Claude Sonnet 4.6 | 23.43 | #90 |
| 15 | Gemini 2.5 Flash Lite | 23.36 | #422 |
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-translation-to · How It Works · Data refreshed daily, snapshot 2026-10-20.