HELM Arabic - Arabic Exams - Social: leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 41 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.3 70B Instruct | 55.51 |
| 2 | Qwen 2.5 72B Instruct | 51.1 |
| 3 | Claude Haiku 4.5 (20251001) | 50.74 |
| 4 | Gemini 2.5 Flash (Thinking) | 50.74 |
| 5 | Llama 4 Maverick Instruct FP8 | 50.37 |
| 6 | Llama 4 Scout Instruct | 50 |
| 7 | GPT-4.1 (2025-04-14) | 50 |
| 8 | Claude Opus 4.7 | 50 |
| 9 | Mistral Large 3 | 49.63 |
| 10 | Gemini 2.5 Flash Lite (Thinking) | 49.63 |
| 11 | GPT-5.4 (2026-03-05) | 49.26 |
| 12 | DeepSeek V3.1 | 48.9 |
| 13 | Claude Sonnet 4.6 | 48.53 |
| 14 | GPT-4.1 Mini | 47.79 |
| 15 | Gemma 4 31B (IT) | 47.79 |
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-arabic-exams-social · How It Works · Data refreshed daily, snapshot 2026-09-19.