ALARB: leaderboard
Arabic Legal Argument Reasoning Benchmark for evaluating LLM reasoning over Saudi commercial court cases and legal arguments.
Metric: Facts+regulations correct (imported). Source: arxiv.org. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 46 | #343 |
| 2 | Qwen 3 14B | 44.6 | #537 |
| 3 | aya-expanse-32B | 36.3 | #692 |
| 4 | Qwen 3 8B | 32.2 | #674 |
| 5 | Gemma 3 12B (IT) | 29.6 | #664 |
| 6 | AceGPT-v2-8B-Chat | 28.9 | #736 |
| 7 | O4 Mini | 27.6 | #180 |
| 8 | AceGPT-v2-32B-Chat | 25.1 | #507 |
| 9 | aya-expanse-8B | 24.6 | #931 |
| 10 | Gemma 3 4B (IT) | 24.5 | #992 |
| 11 | ALLaM-7B-Instruct-preview | 17.2 | #718 |
| 12 | Falcon3-7B-Instruct | 8.7 | #1305 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=alarb · How It Works · Data refreshed daily, snapshot 2026-10-20.