Evals for Every Language - Hebrew (he): leaderboard
Average score across evaluation tasks in Hebrew (he), from Evals for Every Language.
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Around January 2027. 67 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 | 80.81 | #142 |
| 2 | Gemini 3.1 Pro (Preview) | 79.97 | #58 |
| 3 | Claude Opus 4.7 | 78.66 | #49 |
| 4 | Claude Sonnet 4.6 | 77.97 | #90 |
| 5 | Gemini 3.1 Flash Lite | 77.53 | #191 |
| 6 | GPT-5.1 | 76.82 | #135 |
| 7 | Claude Sonnet 4 | 76.74 | #199 |
| 8 | Gemini 2.5 Pro | 76.39 | #152 |
| 9 | DeepSeek V3.1 | 75.78 | #264 |
| 10 | Grok 4.20 | 75.78 | #175 |
| 11 | GPT-5.3 | 75.64 | #103 |
| 12 | DeepSeek V3.2 Exp | 75.43 | #234 |
| 13 | Gemma 3 27B (IT) | 75.03 | #523 |
| 14 | Claude Opus 4.8 | 75.02 | #37 |
| 15 | GPT-5.2 | 74.23 | #110 |
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-language-he · How It Works · Data refreshed daily, snapshot 2026-10-20.