jais-adapted-13B-chat: benchmark results
Inception (G42) and Cerebras' Jais Adapted 13B Chat, fine-tuned from their Jais Adapted 13B base model, released under Apache 2.0. Provider: Inception (G42) & Cerebras. Released 2024-08-02. Access: Open.
Aggregate Rating 1457 ± 20, rank #957 of 1654 rated models, from 281 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Arabic LLM - Aratrust Offensive | 94.2 | Accuracy (%) | 93.5 |
| QIMMA - ArabCulture | 72.16 | Accuracy (%, log-likelihood multiple choice) | 90.2 |
| QIMMA - AraDiCE-Culture | 75.56 | Accuracy (%, log-likelihood multiple choice) | 87.8 |
| Open Arabic LLM - Alghafa Multiple Choice Rating Sentiment Task | 54.65 | Accuracy (%) | 83.6 |
| Open Arabic LLM - Arabic MMLU History (Middle School) | 67.98 | Accuracy (%) | 82.1 |
| Open Arabic LLM - Alghafa Multiple Choice Rating Sentiment NO Neutral Task | 79.79 | Accuracy (%) | 80.9 |
| Open Arabic LLM - Arabic MMLU Economics (Middle School) | 77.01 | Accuracy (%) | 67 |
| Open Arabic LLM - Aratrust Illegal | 88.68 | Accuracy (%) | 66.5 |
| HELM Arabic - Arabic Exams - Biology | 40 | Exact Match (%) | 66.2 |
| Open Arabic LLM - Arabic MMLU Computer Science (Middle School) | 85.19 | Accuracy (%) | 66 |
| Open Arabic LLM - Arabic MMLU Social Science (Middle School) | 51.45 | Accuracy (%) | 64.5 |
| Open Arabic LLM - Aratrust Ethics | 71.67 | Accuracy (%) | 63.7 |
Interactive version: theaggregate.ai/model?slug=jais-adapted-13b-chat · How It Works · Data refreshed daily, snapshot 2026-10-24.