HELM Arabic Enterprise - Arabic Finance Bool: leaderboard
Metric: EM. Source: crfm.stanford.edu. 35 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 4 31B (IT) | 82.46 |
| 2 | Claude Opus 4.7 | 78.95 |
| 3 | GPT-5.4 (2026-03-05) | 75.44 |
| 4 | Qwen3 235B A22B Instruct 2507 FP8 | 71.93 |
| 5 | Claude Sonnet 4.6 | 70.18 |
| 6 | DeepSeek V4 Pro | 70.18 |
| 7 | Llama 3.3 70B Instruct | 68.42 |
| 8 | Qwen 3.5 397B A17B | 63.16 |
| 9 | Mistral Large 3 | 61.4 |
| 10 | Mistral Large 2 (Nov) Instruct (2411) | 61.4 |
| 11 | Gemini 2.5 Flash (Thinking) | 59.65 |
| 12 | Gemini 2.5 Flash Lite (Thinking) | 59.65 |
| 13 | ALLaM-7B-Instruct-preview | 59.65 |
| 14 | DeepSeek V3.1 | 57.89 |
| 15 | Llama 4 Scout Instruct | 57.89 |
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-enterprise-arabic-finance-bool · How It Works · Data refreshed daily, snapshot 2026-09-08.