BloomBench - Arabic: leaderboard
BloomBench (Almieyar-Oryx series, QCRI, 2026) is a bilingual English and Arabic multiple-choice benchmark for vision-language models organized by the six levels of the revised Bloom taxonomy, scored as zero-shot accuracy in percent from the answer letter parsed from generated text (chance 25). This board is micro accuracy over all 7,747 Modern Standard Arabic items; higher is better.
Metric: Micro accuracy (%) on all 7,747 Modern Standard Arabic BloomBench items across the six Bloom levels, four-option image-question multiple choice, zero-shot, temperature 0, the chosen letter parsed from the generated answer (Regex-based Answer Extraction; unparseable answers count as wrong); chance 25; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 4 31B | 87.6 | #394 |
| 2 | Gemma 3 27B | 85.9 | #1631 |
| 3 | Gemma 4 26B A4B | 84.6 | #615 |
| 4 | Gemma 3 12B | 83.6 | #2000 |
| 5 | Qwen 2.5 VL 7B | 79.2 | #1787 |
| 6 | Qwen 2 VL 7B | 77.3 | #2261 |
| 7 | GPT-4o Mini | 76.9 | #1623 |
| 8 | Gemma 3 4B | 72.9 | #5064 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=bloombench-arabic · How It Works · Data refreshed daily, snapshot 2026-11-13.