Mizan - MSA: leaderboard
The Modern Standard Arabic track of Mizan, which most of the field saturates at 100 on the frozen pilot board. Auto-scored; higher is better. Source: arxiv.org/abs/2609.13980.
Metric: Score (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 27 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 100 | #357 |
| 2 | Gemini 2.5 Pro | 100 | #159 |
| 3 | GPT-5 | 100 | #105 |
| 4 | Gemini 2.5 Flash | 100 | #255 |
| 5 | Gemma 4 31B (IT) | 100 | #178 |
| 6 | Claude Opus 5 | 100 | #16 |
| 7 | GPT-6 Astra | 100 | #6 |
| 8 | Claude Sonnet 5 | 100 | #65 |
| 9 | Claude Fable 5 | 100 | #20 |
| 10 | Claude Sonnet 4.5 | 97.5 | #148 |
| 11 | GPT-OSS-120B | 97.5 | #346 |
| 12 | Claude Haiku 4.5 | 97.5 | #287 |
| 13 | Gemma 3 27B (IT) | 95 | #528 |
| 14 | Llama 3.3 70B Instruct | 95 | #541 |
| 15 | DeepSeek V3.2 | 95 | #210 |
No result here: #3 Claude Opus 5.5, #7 GPT-6.1 Sol, #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=mizan-msa · How It Works · Data refreshed daily, snapshot 2026-11-05.