HELM Arabic Enterprise - Arabic Finance Calculation: leaderboard
Metric: Accuracy. Source: crfm.stanford.edu. 35 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 | 75.34 |
| 2 | Claude Sonnet 4.6 | 64.84 |
| 3 | DeepSeek V4 Pro | 63.47 |
| 4 | Qwen3 235B A22B Instruct 2507 FP8 | 63.01 |
| 5 | GPT-5.4 (2026-03-05) | 62.56 |
| 6 | Llama 4 Maverick Instruct | 60.27 |
| 7 | Gemma 4 31B (IT) | 59.36 |
| 8 | DeepSeek V3.1 | 57.99 |
| 9 | Qwen 3.5 397B A17B | 55.25 |
| 10 | Gemini 2.5 Flash (Thinking) | 54.79 |
| 11 | Mistral Large 3 | 53.88 |
| 12 | Claude Haiku 4.5 (20251001) | 52.51 |
| 13 | Gemini 2.5 Flash Lite (Thinking) | 50.68 |
| 14 | Qwen 3.5 9B | 49.77 |
| 15 | Llama 4 Scout Instruct | 46.12 |
Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-enterprise-arabic-finance-calculation · How It Works · Data refreshed daily, snapshot 2026-09-08.