ArabCulture-Dialogue - Dialect to MSA Translation (BLEU): leaderboard
ArabCulture-Dialogue (MBZUAI, 2026) asks models to translate culturally grounded dialogues from an Arab country dialect into Modern Standard Arabic, zero-shot with the target country and region named in the prompt. The score is BLEU-4 against the human reference translation on a 0-100 scale; higher is better.
Metric: BLEU-4 (0-100, the paper's 0-1 score times 100) against the human reference, zero-shot Dialect to MSA translation of the dialogues of ArabCulture-Dialogue (13 Arab countries, parallel Modern Standard Arabic and country-dialect versions of 3,471 culturally grounded multi-turn dialogues built from ArabCulture), with the target country and region named in the prompt; higher is better. Source: arxiv.org. Saturation forecast: Around March 2028. 11 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 43.4 | #178 |
| 2 | Gemini 2.5 Pro | 43 | #276 |
| 3 | ALLaM-7B-Instruct-preview | 40.5 | #2001 |
| 4 | Fanar-1-9B | 36.7 | #1913 |
| 5 | Gemma 2 9B (IT) | 31.9 | #2369 |
| 6 | Hala-9B | 31.5 | #1162 |
| 7 | c4ai-command-r7B-arabic-02-2025 | 31 | #2135 |
| 8 | Qwen 3 8B | 26.5 | #1873 |
| 9 | Llama 3.1 8B Instruct | 24.2 | #4600 |
| 10 | SILMA-9B-Instruct-v1.0 | 21.8 | #3847 |
| 11 | jais-adapted-7B-chat | 17.8 | #5831 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=arabculture-dialogue-dialect-to-msa-translation-bleu · How It Works · Data refreshed daily, snapshot 2026-11-21.