Arabic Broad Leaderboard - Translation (incl Dialects): leaderboard
Translation involving Arabic, including dialects: the 36-question Translation (incl Dialects) category of SILMA's Arabic Broad Benchmark, whose answers are scored 0-10 by rules or an LLM judge. Category average score (0-10), higher is better; each model's latest result file is used.
Metric: Average Score (0-10). Source: huggingface.co. Saturation forecast: Estimated already saturated. 121 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 8.39 | #357 |
| 2 | GPT-5.2 | 8.17 | #115 |
| 3 | GPT-5.5 | 8.06 | #33 |
| 4 | Gemini 3.5 Flash | 8.06 | #66 |
| 5 | GPT-5.6 Pro Sol | 8.06 | #21 |
| 6 | GPT-6.1 Pro Sol | 8.06 | #11 |
| 7 | Gemini 4 Argon | 8 | #9 |
| 8 | DeepSeek V4.1 Flash | 7.97 | #37 |
| 9 | Muse Spark 1.1 | 7.97 | #55 |
| 10 | Gemini 3.1 Pro (Preview) | 7.94 | #64 |
| 11 | GPT-5 | 7.92 | #105 |
| 12 | Claude Sonnet 4 (20250514) | 7.92 | #223 |
| 13 | Claude Opus 4.7 | 7.89 | #51 |
| 14 | Gemma 4 26B A4B (IT) | 7.89 | #293 |
| 15 | Gemini 3.8 Flash | 7.89 | #30 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=arabic-broad-leaderboard-translation-incl-dialects · How It Works · Data refreshed daily, snapshot 2026-11-05.