jais-adapted-70B-chat: benchmark results
Inception (G42) and Cerebras' Jais Adapted 70B Chat, fine-tuned from their Jais Adapted 70B base model, released under Apache 2.0. Provider: Inception (G42) & Cerebras. Released 2024-08-02. Access: Open.
Aggregate Rating 1543 ± 22, rank #627 of 1635 rated models, from 273 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Arabic LLM - Alghafa Multiple Choice Rating Sentiment NO Neutral Task | 84.35 | Accuracy (%) | 97.5 |
| Open Arabic LLM - Arabic MMLU Economics (Middle School) | 82.76 | Accuracy (%) | 96.6 |
| QIMMA - ArabCulture | 78 | Accuracy (%, log-likelihood multiple choice) | 96.3 |
| Open Arabic LLM - Alghafa Multiple Choice Sentiment Task | 43.6 | Accuracy (%) | 96 |
| Open Arabic LLM - Arabic MMLU History (Middle School) | 73.89 | Accuracy (%) | 96 |
| Open Arabic LLM - Alghafa Multiple Choice Rating Sentiment Task | 57.05 | Accuracy (%) | 95.1 |
| HELM Arabic - Alghafa - Multiple Choice Facts Truefalse Balanced Task | 96 | Exact Match (%) | 92.5 |
| QIMMA - AraDiCE-Culture | 76.67 | Accuracy (%, log-likelihood multiple choice) | 89.6 |
| Open Arabic LLM - Arabic MMLU HT Virology | 51.2 | Accuracy (%) | 88.6 |
| Open Arabic LLM - Arabic MMLU Philosophy (High School) | 69.23 | Accuracy (%) | 88.6 |
| Open Arabic LLM - Arabic MMLU Islamic Studies | 62.91 | Accuracy (%) | 87.7 |
| Open Arabic LLM - Aratrust MentalHealth | 94.74 | Accuracy (%) | 87.3 |
Interactive version: theaggregate.ai/model?slug=jais-adapted-70b-chat · How It Works · Data refreshed daily, snapshot 2026-10-15.